Skip to main content

Full text of "Encyclopaedia Dictionary Islam Muslim World, etc, Gibb, Kramer, scholars. 13 vols & 12 vols. 1960-2004.1875.2009."

See other formats







and the late G. LECOMTE 








Members: C.E. Bosworth, J.T.P. de Bruijn, A. Dias Farinha, E. van Donzel, J. van Ess, W.P. Heinrichs, 

RJ. Kasteleijn, A.K.S. Lambton, the late G. Lecomte, B. Lewis, R. Mantran, F. Meier, 

F. Rosenthal, F. Rundgren, A.L. Udovitch. 

Associated members: Haul Inalck, S.H. Nasr, M. Talbi. 

The preparation of this volume of the Encyclopaedia of Islam was made possible 
in part through grants from the Research Tools Program of the National Endowment 
for the Humanities, an independent Federal Agency of the United States Government; 
the British Academy; the Oriental Institute, Leiden; Academie des Inscriptions et 
Belles-Lettres; and the Royal Netherlands Academy of Sciences. 

The articles in this volume were published in double fascicules of 1 1 2 pages, the dates of publication being: 
1995: Fascs. 147-150, pp. 1-208 1996: Fascs. 151-157, pp. 209-600 

1997: Fascs. 158-162, pp. 601-920 

ISBN 90 04 10422 4 

© Copyright 1997 by Konmklijke Brill, Leiden, The Netherlands 

All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval 

system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording 

or otherwise, without prior written permission of the publishers. 

Authorization to photocopy items for internal or personal use is granted by Brill provided that 

the appropriate fees are paid directly to Copyright Clearance Center, 222 Rosewood Drive, Suite 910, 

Darners, MA 01923, USA. Fees are subject to change. 



For the benefit of readers who may wish to follow up an individual contributor's articles, the Editors have 
decided to list after each contributor's name the pages on which his signature appears. Academic but not 
other addresses are given (for a retired scholar, the place of his last known academic appointment). 

In this list, names in square brackets are those of authors of articles reprinted or revised from the first 
edition of this Encyclopaedia or from the Shorter Encyclopaedia of Islam. An asterisk after the name of the author 
in the text denotes an article reprinted from the first edition which has been brought up to date by the 
Editorial Committee; where an article has been revised by a second author his name appears within square 
brackets after the name of the original author. 

Virginia Aksan, McMaster University, Hamilton, On- 
tario. 134, 414 
Hamid Algar, University of California, Berkeley. Ill, 

T.T. Allsen, Trenton State College. 44 
Edith G. Ambros, University of Vienna. 354, 896 
R. Amitai-Preiss, Hebrew University, Jerusalem. 739 
L. Ammann, University of Freiburg. 176 
Salman H. Al-Ani, Indiana University, Bloomington. 

Sarah Ansari, University of London. 636 
A. Arazi, Hebrew University, Jerusalem. 289, 303, 

389, 462, 777, 803, 868 
[C. van Arendonk, Leiden]. 905 
R. Arnaldez, University of Paris. 382, 806 
W. Arnold, University of Heidelberg. 439 
J.P. Asmussen, University of Copenhagen. 211 
Sarab Atassi, Institut Francais d'Etudes Arabes, 

Damascus. 795 
Khalil 'Athamina, Birzeit University. 250 
M. Athar Ali, Aligarh Muslim University. 181, 193, 

195, 196, 339, 738, 846, 909 
A. Ayalon, Tel Aviv University. 152, 506 
Maha Azzam, London. 746 
Roswitha Badry, University of Freiburg. 481 
C. Bailey, Tel Aviv University. 625 
M.A. al-Bakhit, Al al-Bayt University, Mafraq, Jordan. 

211, 374 
CifiDEM Balim, University of Manchester. 469 
Amatzia Baram, University of Haifa. 369 
R.B. Barnett, University of Virginia, Charlottesville. 

A. Bazzana, University of Lyons. 260 
Anna Lwia Beelaert, Leiden. 1 16 
the late A.F.L. Beeston, Oxford. 151 
M.A.J. Beg, Cambridge. 288, 666 
Fethi Bejaoui, National Institute of Archaeology, 

Tunis. 740 
A. Ben Abdesselem, Institut National des Langues 

et Civilisations Orientales, Paris. 219, 233, 350, 

[Moh. Ben Cheneb, Algiers]. 290 
O. Benchekh, Centre National de la Recherche Scien- 

tifique, Paris. 306 
H. Bencheneb, Paris. 22, 348, 538 
Th. Bianquis, University of Lyons. 110, 392, 789 
A. Bj0Rkelo, University of Bergen. 301 
[W. BjOrkman, Uppsala]. 678 

Sheila S. Blair, Richmond, New Hampshire. 47, 861 
F.C. de Blois, Royal Asiatic Society, London. 1 1 2, 

191, 209, 212, 220, 338, 495, 538, 615, 616, 

C.E. Bosworth, University of Manchester. 7, 9, 11, 

17, 28, 34, 39, 49, 51, 54, 69, 70, 87, 93, 98, 

114, 116, 121, 158, 159, 165, 170, 195, 197, 219, 

255, 262, 273, 275, 309, 311, 315, 340 440, 442, 

488, 489, 490, 491, 496, 499, 501, 505, 508, 513, 
558, 574, 575, 576, 582, 592, 612, 614, 617, 622, 
635, 657, 660, 667, 668, 673, 685, 692, 693, 694, 
702, 736, 737, 740, 745, 763, 766, 773, 819, 822, 
843, 851, 854, 859, 860, 869, 872, 893, 894, 899, 
913, 919 

A. Boudot-Lamotte, Institut National des Langues et 
Civilisations Orientales, Paris. 380 

G. BOwering, Yale University. 812 

W.C. Brice, University of Manchester. 603 

D.W. Brown, Mount Holyoke College, South Hadley, 

MA. 881 
J.T.P. de Bruijn, University of Leiden. 5, 212, 221, 

239, 298, 415, 434, 467 
M.M. van Bruinessen, University of Utrecht. 153, 

246, 283 
[V.F. Buchner]. 311, 679 
R.W. Bulliet, Columbia University. 400 
Kathleen R.F. Burrill, Columbia University. 59, 

150, 419, 420, 762, 843, 877 
P. Cachia, Columbia University. 353, 483 
N. Calder, University of Manchester. 36, 326 
Y. Callot, University of Tours. 368 
J. Calmard, Centre National de la Recherche Scienti- 

fique, Paris. 192 
Sheila R. Canby, British Museum, London. 692 
J. Carswell, Sotheby's, London. 417, 648 
M.G. Carter, University of Oslo. 360, 531 
J.-C. Chabrier, Centre National de la Recherche Scien- 

tifique, Paris. 20 

E. Chaumont, Centre National de la Recherche Scien- 
tifique, Aix-en-Provence. 185, 189, 394, 483, 813, 

the late ]. Chelhod, Paris. 113 

Moncef Chenoufi, University of Tunis. 565 

V. Christides, University of Ioannina. 445 

Nathalie Clayer, Ecole des Hautes Etudes en 

Sciences Sociales, Paris. 156, 300, 876 
J. Cler, University of Strasbourg. 121 
W.L. Cleveland, Simon Fraser University, British 

Columbia. 248 
[G.S. Colin, Paris]. 375, 734 
J. Couland, University of Paris. 522 
Patricia Crone, Princeton University. 825 
Yolande Crowe, Geneva. 638 

F. Daftary, Institute of Ismaili Studies, London. 114, 
296, 435 

H. Daiber, University of Frankfurt. 247 

B. van Dalen, Kyoto Sangyo University. 294 
M.W. Daly, University of Tennessee, Chattanooga. 

R.E. Darley-Doran, Winchester. 203, 599 

G. David, Eotvos Lorand University, Budapest. 557, 
632, 920 

J.-C. David, Centre National de la Recherche Scien- 
tifique, Lyons. 796 

R.H. Davison, George Washington University. 303 
J.-Ch. Depaule, Centre National de la Recherche 

Scientifique, Paris. 321 
A. Dietrich, University of Gottingen. 9, 101, 249, 

432, 435, 496, 615, 910 
Moktar Djebli, University of Tunis. 114, 343, 472, 


A. Dubinski, University of Warsaw. 135 
R. Eisener, University of Bamberg. 822 
Taieb El Acheche, University of Tunis. 762, 770 
A. Elad, Hebrew University, Jerusalem. 573 
W. Ende, University of Freiburg im Breisgau. 217, 

261, 315 
S. Enderwitz, Free University, Berlin. 516 
C. Ernst, University of North Carolina, Chapel Hill. 

362, 483 
T. Fahd, University of Strasbourg. 6, 85, 97, 169, 

228, 281, 291, 314, 382, 408, 424, 440, 571, 613, 

738, 763, 892, 909 
[H.G. Farmer, Glasgow]. 11 
Suraiya Faroq_hi, University of Munich. 126, 200, 

412, 545, 575, 656, 691, 800 
P.-B. Fenton, University of Strasbourg. 160, 824 
Halima Ferhat, University of Rabat. 190, 259 
Maribel Fierro, C.S.I.C, Madrid. 19, 38, 165, 258, 

306, 340, 365, 513 
Kais M. Firro, University of Haifa. 494 
W. Floor, Bethesda, Maryland. 646 
J. Fontaine, Institut des Belles Lettres Arabes, Tunis. 

G.S.P. Freeman-Grenville, SherifF Hutton, York. 150, 

479, 504, 535, 600, 685, 686, 702, 716 
Y. Frenkel, University of Haifa. 765 

C. Fyfe, University of Edinburgh. 551 
H. Gaube, University of Tubingen. 796 
[M. Gaudefroy-Demombynes, Paris]. 391 

GJ.H. van Gelder, University of Groningen. 27, 282 
E. Geoffroy, University of Strasbourg. 398, 916 
A. Ghabin, Hebrew University, Jerusalem. 629 
A. Ghedira, University of Lyons. 161 
Alma Giese, Cambridge, Mass. 497 
Cl. Giixiot, University of Aix-en-Provence. 320, 372, 
607, 609 

D. Gimaret, Ecole Pratique des Hautes Etudes, Paris. 
179, 201, 203, 409, 486, 552, 614, 743 

M. Glunz, University of Washington, Seattle. 282 

Fatma Muge Gocek, University of Michigan. 416 

P.B. Golden, Rutgers University. 115, 533 

W.A. Graham, Harvard University. 337 

Vincenza Grassi, University of Naples. 591, 670 

[A. Grohmann, Vienna]. 55 

A.H. de Groot, University of Leiden. 15, 24, 61, 

698, 829, 845 
P. Guichard, University of Lyons. 352, 789, 791 
J.-P. Guillaume, University of Paris. 102, 640 
J.GJ. ter Haar, University of Leiden. 329 
C.P. Haase, University of Kiel. 113, 644, 872 
[T.W. Haig, London]. 26, 50, 658 
W. Hale, University of London. 524 
Wael B. Hallaq, McGill University. 359 
H. Halm, University of Tubingen. 299, 357, 361, 501, 

685, 686 
Talat Sait Halman, New York University. 150, 213, 

240, 415, 444, 738 
A. Hamdani, University of Wisconsin, Milwaukee. 895 
A. Hamori, Princeton University. 218 
Shai Har-El, Northbrook, Illinois. 610 
Angelika Hartmann, University of Wiirzburg. 782 

J.A. Haywood, Lewes, East Sussex. 90, 378, 434, 

470, 895, 916 
P. Heath, Washington University, St. Louis. 665 
[W. Heffening, Cologne]. 63 
S. Heidemann, University of Jena. 894 
W.P. Heinrichs, Harvard University. 56, 355, 442, 

C.J. Heywood, University of London. 693 
the late D.R. Hill, London. 3 
Enid Hill, American University in Cairo. 19 
Carole Hillenbrand, University of Edinburgh. 193 
M.C. Hillmann, University of Texas, Austin. 658 
the late M. Hiskett, London. 244, 470 
R. Hitchcock, University of Exeter. 534 
A. Hofheinz, University of Bergen. 89 
J.P. Hogendijk, University of Utrecht. 481 
P.M. Holt, Oxford. 155, 181 

M.B. Hooker, Australian National University, Can- 
berra. 195, 328 
Virginia Matheson Hooker, Australian National Uni- 
versity, Canberra. 41, 245 
Genevieve Humbert, Institut de Recherche et d'His- 

toire des Textes, Paris. 669 
J.O. Hunwick, Northwestern University. 730 
C. Imber, University of Manchester. 60, 844 
Haul Inalcik, Bilkent University, Ankara. 131 
Mehmet Ip§irli, University of Istanbul. 127 
[Muhammad Iq^bal, Lahore]. 581 
R. Israeli, Hebrew University, Jerusalem. 625 
Svetlana Ivanova, National Library, Sofia. 504, 

M.Y. Izzi Dien, University of Wales, Lampeter. 94, 

248, 349, 471, 495, 770, 806 
P.Jackson, University of Keele. 112, 159, 199, 283 
S.A. Jackson, Indiana University, Bloomington. 436 
J.J.G. Jansen, University of Leiden. 1 18, 378 
O. Jastrow, University of Heidelberg. 376, 575 
Mohamed Jedidi, University of Tunis. 902 
A.H. Johns, Australian National University, Can- 
berra. 296 
Penelope Johnstone, University of Oxford. 873, 902 
G.H.A. Juynboll, The Hague. 163, 249, 492, 567, 

762, 874, 881 
Wadad Kadi, University of Chicago. 1 1 7 
O. Kahl, Rochdale, Lanes. 8, 667, 898 
Ousmane Kane, University of Saint-Louis, Senegal. 

244, 445, 853 
A.S. Kaye, California State University, Fullerton. 517, 

752, 761 
Monique Kervran, University of Paris. 776 
M. Khadduri, Johns Hopkins University, Baltimore, 

MD. 846 
Ansar Zahid Khan, Pakistan Historical Society, 

Karachi. 440 
Zafarul-Islam Khan, Institute of Islamic and Arabic 

Studies, New Delhi. 378 
R.G. Khoury, University of Heidelberg. 169 
M. Kiel, University of Utrecht. 137 
D.A. King, University of Frankfurt. 180, 253 
A. Knysh, University of Michigan, Ann Arbor. 288, 

Ebba Koch, University of Vienna. 64, 736 
E. Kohlberg, Hebrew University, Jerusalem. 207, 210, 

509, 827 
J. Kostiner, Tel Aviv University. 908 
[J.H. Kramers, Leiden]. 64, 68, 654, 706, 712, 802, 

K. Kreiser, University of Bamberg. 855 
Remke Kruk, University of Leiden. 550 
Ann Kumar, Australian National University, Canberra. 

P. Kunitzsch, University of Munich. 472 
M. Kunt, University of Cambridge. 13 
Mohamed Mokhtar Labidi, University of Tunis. 396 
P.L. Lamant, Institut National des Langues et Civi- 
lisations Orientales, Paris. 18 
Ann K.S. Lambton, Kirknewton, Northumberland. 

438, 479, 734 
Fidelity Lancaster, Orkney. 317, 376, 815 
W. Lancaster, Orkney. 317, 376, 815 
D.M. Last, University College, London. 712 
M. Lavergne, Centre National de la Recherche Scien- 

ufique, Paris. 100, 885 
M. Lecker, Hebrew University, Jerusalem. 556, 818 
S. Leder, University of Halle. 65, 164, 805, 848 
J.L. Lee, Leeds University. 431 
Cl. Lefebure, Centre National de la Recherche Scien- 

tifique, Paris. 901 
A. Leguil, Institut National des Langues et Civilisations 

Orientales, Paris. 689 
G. Leiser, Vacaville, California. 62, 410, 826 
J. Lentin, University of Paris. 281 
[E. Levi-Provencal, Paris]. 255, 256, 317, 353, 357, 

374, 446, 534 
K. Lewinstein, Smith College, Northampton, MA. 

L. Lewisohn, University of London. 377, 484, 863 
Chang-Kuan Lin, National Cheng-chi University, 

Taipei. 416, 650, 920 
D.P. Little, McGill University. 494, 539 
P. Lory, Ecole Pratique des Hautes Etudes, Paris. 66, 

172, 175, 220 
R.D. McChesney, New York University. 428, 431 

D. MacEoin, University of Durham. 64, 380, 405, 

W. Madelung, University of Oxford. 314, 424, 767, 

824, 875 
Beatrice Forbes Manz, Tufts University, Medford, 

Mass. 198 
Manuela Marin, University of Madrid. 306, 340, 

364, 441, 506, 512, 741 
U. Marzolph, Enzyklopadie des Marchens, Gottingen. 

[L. Massignon, Paris]. 159 
R. Matthee, University of Delaware. 821, 855 
C.P. Melville, University of Cambridge. 49 
Rachel Milstein, Hebrew University, Jerusalem. 295 
[V. Minorsky, Cambridge]. 65, 246, 500, 658, 681, 

727, 831, 832, 878 
M. Mokri, Centre National de la Recherche Scien- 

tifique, Paris. 643, 857 
J.-P. Molenat, Centre National de la Recherche 

Scientifique, Paris. 6, 255, 257, 258, 304, 305, 308 
L. Molina, University of Granada. 871 
G. Monnot, Ecole Pratique des Hautes Etudes, Paris. 

216, 671, 870 
J.E. Montgomery, University of Leeds. 9, 804, 892 
S. Moreh, Hebrew University, Jerusalem. 230, 370, 

443, 465, 498 
M. Morony, University of California, Los Angeles. 83 
D.W. Morray, University College Dublin. 154, 611, 

702, 910 
Ziba Moshaver, University of London. 524 
H. Motzki, University of Nijmegen. 8 
J.-M. Mouton, University of Picardy, Amiens. 41 1 
W.W. Muller, University of Marburg. 166 
R. Murphey, University of Birmingham. 743, 843 
Azim Nanji, University of Florida, Gainesville. 85 
I.R. Netton, University of Leeds. 247, 251, 360, 694, 


E. Neubauer, University of Frankfurt. 101, 351, 361, 

Angelika Neuwirth, Orient-Institut, Beirut. 366 
J.S. Nielsen, Selly Oak Colleges, Birmingham. 510 
K.A. Nizami, Aligarh University. 85, 120, 196, 356, 

370, 735, 801 
G. Nonnemann, Lancaster University. 349 
H.T. Norris, University of London. 446 

A. Northedge, University of Paris. 883 

Linda S. Northrup, University of Toronto. 433 
Ozdemir Nutku, University of Izmir. 646 
R.S. O'Fahey, University of Bergen. 406 

B. O'Kane, American University of Cairo. 630 
G. Oman, University of Naples. 590, 809 

M. Orwin, University of London. 714, 715, 718, 727 

G. Osman, University of Naples. 50 

J.P. Pascual, Centre National de la Recherche Scien- 
tifique, Aix-en-Provence. 795 

V. Perthes, Research Institute for International Politics 
and Security, Ebenhausen. 277 

R. Peters, University of Amsterdam. 208 

Ch. Picard, Societe des Medievistes, St.-Etienne. 307 

D. Pingree, Brown University. 641 

[M. Plessner, Jerusalem]. 778 

I. Poonawala, University of California, Los Angeles. 

201, 221, 398, 825, 829 

A. Popovic, Centre National de la Recherche Scienti- 
fique, Paris. 34, 681 

C. Poujol, Institut National des Langues et Civilisations 
Orientales, Paris. 660 

L. Pouzet, Saint-Joseph University, Beirut. 406 
Nasser Rabbat, Massachusetts Institute of Technology, 

Cambridge. 176 
H.P. Raddatz, Hallgarten. 772 
Munibur Rahman, Oakland University, Rochester, 

Michigan. 162, 214, 242, 379, 396, 397, 402, 436, 

437, 510, 897 
M. al-Rasheed, King's College, London. 291 
W. Raven, University of Frankfurt. 663 

A. Raymond, University of Aix-en-Provence. 169, 194, 
645, 792 

B. Reinert, University of Zurich. 59, 313, 873 
A.K. Reinhart, Dartmouth College, Hanover, NH. 

251, 313, 498, 868 
R.C. Repp, University of Oxford. 402, 874 
D.F. Reynolds, University of California, Santa Bar- 
bara. 236 
D.S. Richards, University of Oxford. 373, 487 
M.J.E. Richardson, University of Manchester. 149 
A. Rippin, University of Calgary. 27, 300, 383, 409, 

491, 535, 547, 550 
Ch. Robin, Centre National de la Recherche Scienti- 
fique, Aix-en-Provence. 92, 425, 426, 676 
C.F. Robinson, University of Oxford. 165, 202 
Ruth Roded, Hebrew University, Jerusalem. 90 
F. Roldan-Castro, University of Seville. 177, 350 
F. Rosenthal, Yale University, New Haven. 35, 152, 

202, 368, 403, 552 

A. Rouaud, Centre National de la Recherche Scienti- 
fique, Paris. 399, 412, 425, 426, 722 

E.K Rowson, University of Pennsylvania, Philadelphia. 
493, 611 

J. Sadan, Tel Aviv University. 312 

P.C. Sadgrove, University of Manchester. 502 

T. Saguchi, Kanazawa. 304, 441 

K.S. Salibi, American University of Beirut. 435 

Dj. Sari, Algiers. 358 

A. Sawides, Centre for Byzantine Studies, Athens. 
20, 137, 419, 484 

[H.H. Schader]. 87 

R.P. Scheindlin, The Jewish Theological Seminary of 
America, New York. 343 

Annemarie Schimmel, Bonn. 179, 298 

J. Schmidt, The Hague. 697 

Irene Schneider, University of Cologne. 548 

Gudrun Schubert, University of Basel. 859 

R. Schulze, University of Bern. 913 

O. Schumann, University of Hamburg. 814, 853, 877, 

T. Seidensticker, University of Jena. 282 
R. Sellheim, University of Frankfurt. 41 1 
C. Shackle, University of London. 637 
Irfan Shahid, Georgetown University, Washington, 

D.C. 28, 425 
P. Shinar, Hebrew University, Jerusalem. 764, 858 
A. Sidarus, University of Evora. 309, 849 
Mohammad Yusuf Siddiq, Islamic University Kushtia, 

Bangladesh. 84, 576, 611, 728, 877 
I.H. Siddiqui, Aligarh Muslim University. 297, 448, 667 
M.-Cl. Simeone-Senelle, Centre National de la 

Recherche Scientifique, Paris. 8 1 1 
Elizabeth M. Sirriyeh, University of Leeds. 905 
G.R. Smith, University of Manchester. 3, 115, 201, 

338, 356, 439, 739, 817 
S.C. Smith, University of London. 647 
P. Smoor, University of Amsterdam. 339 
F. Sobieroj, University of Giessen. 433, 778, 786 
S. Soucek, Princeton University. 136, 149, 536, 680, 

J.L. Spauldino, Kean College, Union, New Jersey. 

651, 698 
Susan A. Spectorsky, City University of New York. 

N.A. Stillman, University of Oklahoma, Norman. 769 
W. Stoetzer, University of Leiden. 55, 113 
M.E. Subtelny, University of Toronto. 68 
R. Talmon, University of Haifa. 551, 603 
R. Tapper, University of London. 225 
A.I. Tayob, University of Cape Town. 731 
Gc-NtiL Alp ay Tekin, Harvard University. 123, 917 
Abdeljalil Temimi, Zaghouan, Tunisia. 657 
M. Terrasse, Ecole Pratique des Hautes Etudes, Paris. 

G.R. Tibbetts, Oxford. 828 

Shawkat M. Toorawa, University of Mauritius. 576 

Farouk Topan, University of London. 918 

R. Traini, University of Rome. 589 

J.L. Triaud, University of Aix-Marseilles. 23, 26, 122, 

148, 760 

J.-F. Troin, University of Tours. 912 

M.O.H. Ursdjus, University of Heidelberg. 673 

G. Veinstein, Ecole des Hautes Etudes en Sciences 
Sociales, Paris. 711, 842 

J. Vernet, Centre National de la Recherche Scienti- 
fique, Paris. 557 

Ch. de La Veronne, Centre National de la Recherche 
Scientifique, Paris. 18, 508, 820 

C.H.M. Versteeoh, University of Nijmegen. 54, 

M.J. Viouera, University Complutense of Madrid. 38, 

F. Vire, Centre National de la Recherche Scientifique, 
Paris. 40, 51, 99, 504, 653, 811 

F.E. Vooel, Harvard University. 696 

D. Waines, University of Lancaster. 93, 225, 805 
A.T. Welch, Michigan State University, East Lansing. 

[AJ. Wenstnck, Leiden]. 154, 380, 742, 903 
M. Winter, Tel Aviv University. 316, 398 
Christine Woodhead, University of Durham. 132, 

149, 212, 348, 414, 610, 630, 631, 713 
M.E. Yapp, University of London. 448 

E. Yarshater, Encyclopaedia Iranica, New York. 697 
St. Yerasimos, Institut Francais d'Etudes Anatoliennes, 

Istanbul. 414 
Alexandra Yerolympos, Aristotelian University, Thes- 

saloniki. 675 
Elizabeth A. Zachariadou, University of Crete. 69, 

[K.V. Zettersteen, Uppsala]. 295 
Hossein Ziai, University of California, Los Angeles. 

Madeline C. Zilfi, University of Maryland, College 

Park. 413 
EJ. Zurcher, University of Nijmegen. 417, 499 



P. 88 b , KIBLA, ii, add at end: A second treatise of Ibn al-Haytham on the determination of the kibla by 
spherical trigonometry is now published in A. Dallal, Ibn al-Haytham's universal solution for finding the 
direction of the Qibla by calculation, in Arabic Science and Philosophy, v (1995), 145-93. 

Further information on the Persian Mecca-centred world-map discussed in the article al-samt is 
included in the article al-tasa and in the addenda and corrigenda to al-samt. 

P. 807", L UGHZ . add to BibL Shams Anwari-Alhosseyni, Logaz und Mo'amma. Eine Quellenstudie zur Kunstform 
des persischen Ratsels, Berlin 1986. 

P. 1169 b , MAGHNATIS. 2, add at end: The treatise in ms. Berlin Ahlwardt 5811 mentioned on p. 1169b 
is in fact a late copy of the treatise by the Yemeni Sultan al-Ashraf— see al-tasa for this and sev- 
eral other additional sources. The various publications by J. Klaproth and E. Wiedemann listed in 
the bibliography, as well as numerous others, have been reprinted in one volume as Fuat Sezgin 
et alii (eds.), Islamic geography, xv, Reprint of studies on nautical instruments, Frankfurt 1992. 



P. 1027, NASRLDS, in genealogical table, for the date of Muhammad XI (el Chiauito), read (1451-2/1453-5), 
and p. 1027", 1. 7 from foot, for 949/1533-4, read 940/1533-4. 


P. 245 b , SHAKAK, 1. 11,> Idris Bidlrsi, read SJiaraf al-DTn Bidllsf. 

P. 447 b , RASHID RipA, add to BibL, first para.: Makalat al-Shaykh Rashld Rida, ed. Yusuf Husayn Ibish 
and Yusuf Kuzma Khun, 5 vols. Beirut 1994 (with good index). 

P. 869 a , SA'IDA GILANI, 1. 4 from bottom, delete: his wife. 

P. 1056 s , al-SAMT, add at end: In May 1995 a second Persian world-map centred on Mecca came to light 
and is now in a private collection. It is very similar to the first one, which became available for 
study in 1989, but more crudely executed. However, it still bears the paraphernalia for a European-type 
universal inclining sundial. (The sundial on the first, not necessarily of the same kind, is missing.) 
It is clear that both maps were copied from others, the original probably bearing more geographical 
information than either of these. 

The first map is still to be dated ca. 1110/1700 (± 20 years); the second is undoubtedly later 
but bears striking resemblances in calligraphy and appendages (such as screws and feet) to the first, 
if not in general aspect and finish. Both maps are not necessarily from Isfahan, for Khurasanian 
forms of the Persian ordinal numerals are found on them. However, apart from various astrolabes 
by Muhammad Zaman (ca. 1075/1665) of Mashhad, no other astronomical instruments survive 
from northeastern Persia in late-Safawid times. 

The geographical data in al-Khazinfs Sandman %i$, which was thought to have been derived 
from a world-map of the same kind as the first Persian world-map — this was stated in the article 
al-samt — was in fact derived by calculation, as research in April 1995 established. The longitude 
and latitude values were indeed read from a map by or in the tradition of al-BTrunT but the kibla- 
values are inaccurate, not because they were read from a hypothetical Mecca-centred map of al- 
Birunf but because al-Khazini rounded them to the nearest 20' before finding the kibla. To do 
this he apparendy used a kibla-table (with values for each degree of longitude difference from Mecca 
up to 60° and each degree of latitude difference up to 30°), which he mentions in the Z*<$ DUt 
which is not contained in any of the known manuscripts; it can further be shown that this kibla- 
table was carelessly computed. In other words, we must look elsewhere for the inspiration for the 
Persian world-maps. 

The geographical data on both of them (ca. 150 localities on each, the selection being slighdy 
different on the two maps) was taken from a TTmurid geographical table compiled in Kish [q.v.] 
ca. 850/1450, and the anonymous compiler of that table (extant in ms. London, B.L. Or. 7489, 
fols. 53a-58b, where, however, the values of the kibla and distance to Mecca for some 250 local- 
ities originally given to seconds have been rounded to minutes) is certainly a possible candidate. 
But the writings of Habash al-Hasib [q.v.] and al-Blruni [q.v.] particularly with regard to the astro- 
labe labelled muballakh (with a melon-shaped rete), indicate that already in the 3rd/9th and 5th/ 11th 
centuries Muslim astronomers were concerned with mappings preserving distance and direction to 
a central point (see E.S. Kennedy, P. Kunitzsch and R.P. Lorch, The Melon Astrolabe of Habash al- 
Hasib, in ZGAIW [1996]), a subject first broached in Europe in the 16th century and first applied 
to a Mecca-centred map in the early 20th century. (D.A. King) 


P. 102", ATHUR, add to BibL: S. Heidemann, al-'Aqr, das islamische Assur. Ein Beitrag zur histonschen Topographic 
in Mbrdmesopotamien, in Karin Bartl and S.R. Hauser (eds.), Continuity and change in northern Mesopotamia 
from the Hellenistic to the early Islamic period, Procs. of a Colloquium held at the Seminar fur Vorder- 
asiatische Altertumskunde, Freie Universitat Berlin, 6th-9th April, 1994, Berlin 1996, 259-85. 

P. 229 b , DIABAL SAYS, add at end: The Arabic inscription of al-Harith b. Djabala from A.D. 521 men- 
tioned here is, in fact, the third one before Islam and the only historical one; in other inscrip- 


tions, including several of the early Islamic period, going up to the time of al-Walld I, the ancient 
form Usays appears. 

add to BibL M. Abu 1-Faradj Al-'Ush, Inscriptions arabes inedites a Djabal Usays, in Annates archeol. 
de Syrie, xiii (1963), 225-39, Ar. version, 281-91; idem, Us dessins mpestres du Gabal 'Usais, in Syria, 
xli (1964), 291-9; idem, Kitabat gkayr manshurat ft Djabal Usays, in al-Abhath, xvii (1964), 227-316; 
idem, in ibid., xviii (1964), 216-17; A. Grohmann, in Arabische paldographie. II. Teil, Das Schriftwesen. 
Die Lapidarschrift, Graz-Vienna-Cologne 1971, 1-17, Abb. 7 d, Tafel I, 2. 


P. 370 a , SHA'UL, Anwar, 1. 4 from the foot, for the Alliance Francaise read the College A-D Sasson and 
the Alliance Israelite Universelle. 

P. 370 b , 1. 30, for Kissat hayaB wadi 'l-Rqfidayn read Kissat hayaR fl wadi 'l-Rafidayn. 

P. 509 a , SHURAYH, 1. -23, for al-Hakam b. 'Uyayna, read al-Hakam b. 'Utayba. 

P. 693 a , SIYAKAT, add to BibL: Miibahat Kutukoglu, Osmanh beigelerinin dili (diplomatic), Istanbul 1994, passim; 
Said Ozturk, Osmanh arsiv belgeterinde siyakat yazisi ve tarihi gelismi, Istanbul 1 996; cf. also, on Fekete, 
Ismet Binark (ed.), Macar asilli ttirk tarihcisi ve arfivist Lajos Fekete'nin arfivciligimizde yeri, Ankara 1994. 


SAN'A 3 , from ancient times thechieftownofthe 
Yemen [q.v.] and present capital of the unified 
Republic of Yemen. Its present population is reckon- 
ed to be just over half a million. The town is situated 
in the centre of the northern highlands of the Yemen 
atlat. 15° 22' N. and long. 44° 11' E., i.e. about 170 
km/106 miles as the crow flies from the nearest point 
on the Red Sea and 300 km/ 186 miles approximately 
from the Indian Ocean port of Aden [see c adan]. 
San'a' is located at a height above sea level of more 
than 2,200 m/7,216 feet. It is all but surrounded by 
mountains, Djabal Nukum (2,892 m/9,486 feet) in 
the east, at the foot of which the town lies spread, 
Djabal c Ayban (3,194 m/10,476 feet) in the west, the 
highest peak in the immediate vicinity, and the twin 
peaks of al-Nahdayn (2,513 m/8,243 feet) which lie 
due south. Its climate is a temperate one, generally 
very dry and mild. Its rainfall pattern is consistent, 
with maximum falls in March, April, May and July, 
August, September (see Acres, Lewcock and Wilson, 
in Serjeant and Lewcock, SanV, 13-19, in Bibl. 
Pre-Islamk SanW 

Despite the current claim that the name of San c a 3 
is derived from the excellence of its trades and crafts 
(perhaps the feminine form of the Arabic adjective 
asna'), it is highly probable that the name is Sabaic 
and, in keeping with the basic meaning in Sabaic of 
the root sn c , means "well fortified" (A.F.L. Beeston 
et alii, Sabaic dictionary, Louvain and Beirut, 1982, 
143). It is certainly as a military centre that the town 
emerges in the pre-Islamic inscriptions, particularly 
as the headquarters of the Sabaeans for their military 
expeditions southwards against Himyar [q.v.). In- 
scriptions Ja 575, 576, for example, speak of Sabaean 
campaigns launched from San c a 3 {bnl hgrnl sn'w) (A. 
Jamme, Sabaean inscriptions from Mahram Bilqis, 
Baltimore 1962, 64 ff.) and Ja 574, 576, 577 (Jamme, 
60 ff.) announce a triumphant return to San c a= 
C-adyl hgrnl sn'-w) from the wars. Apart from being a 
"town" (hgr), San'a 5 was also a mahram (mhrm) which 
Beeston (San c a>, 37) interprets as "a place to which ac- 
cess is prohibited or restricted, no matter whether for 
religious or for other reasons." The palace of Ghum- 
dan [q.v.] is also mentioned in the pre-Islamic inscrip- 
tions (e.g. J. Ryckmans, La Mancie par hrb, in 
Festschrift Werner Casket, Leiden 1968, 263 (NNAG 
12)) and see SanW, 44. Islamic tradition also reports 
that the mid-6th century Abyssinian King Abrahah 
[q.v.] built a church in San c a> (al-Kalls) (al-Tabarl, i, 
934), for a study of which see SanV, 44-9. 

Unfortunately, the inscriptions do not tell us when 
the site of San'a 3 was first settled, nor do they mention 
the town's ancient name of Azal (al-Hamdanl, 55). 
The above quoted inscriptions and a very large ma- 

jority of the texts mentioning the name of the town are 
all to be dated to the 3rd century A.D. The site must, 
ourse, have been settled a very good deal earlier 
Beeston, in SanV, 36-9). 
Early Islamic and mediaeval SanW 

For the more than two centuries of the early Islamic 
history of San c a 3 , we have little more than a list of the 
governors despatched to the Yemen by the Prophet, 
the Orthodox, Umayyad and c Abbasid caliphs. A 

mprehensive list was attempted in SanW, 53-4, and 
Tables 1, 4, 6, 8 and 11 of c Abd al-Muhsin Mad c adj 
M. al-Mad c adj, The Yemen in early Islam, 9-233/630- 
847, a political history, London 1988, should also be 
consulted. The remainder of the period can be divided 
into six as follows: the Yu'firids, 232-387/847-997 
[q.v.], the Sulayhids, 439-92/1047-99 [q.v.], the 
sultans of Hamdan, 493-569/1099-1173, the Ayyu- 
bids, 569-628/1173-1230 [q.v.], the RasQlids 628- 
783/1230-1381 [q.v.] and a period in which the Zaydl 
Imams controlled San c a>, 783-953/1381-1546 [see 

A detailed history of this period cannot be given 
here, and a few notes will suffice. The Yu c firids, it 
might be noted, were the first Yemeni dynasty to take 
other than a purely local control of any part of the 
Yemen. Descended from Himyar, they regarded 
themselves as their legitimate heirs and moved into 
San'a' from their original territory in Shibam when 
they perceived the weakness of the local c Abbasid 
governors posted there. Their 150 years' rule of 
San'a' came to an end in 387/997, leaving the town 
in anarchy until the arrival of the Fatimid Sulayhids 
in 439/1047. 

Our sources are particularly silent about the 
Sulayhids during their San c a 5 period. The zenith of 
Sulayhid rule came later during the period during 
which they were centred in Dhu Djibla, which the 
dynasty settled in about 480/1087. With the death in 
492/1098 of the Z)a c f Saba> b. Ahmad, San c a> was lost 
to the Sulayhids. 

During the period 492-569/1098-1173 Wa> fell 
under the rule of three families of Hamdan from 
Yam, also Isma c ilis like the Sulayhids: Banu Hatim 
(I), Banu '1-Kubayb and Banu Hatim (II) (G.R. 
Smith, The Ayyubids and early Rasulids in the Yemen, 
London 1974-8, 2 vols., ii, 68, 70-5). Depending al- 
most entirely on the forces which could be mustered 
from Hamdan [q.v.], the three families controlled 
San c a' and were still there to contest authority with 
the Ayyubids after their conquest of Tihama [q.v.] 
and the southern highlands from Egypt in 569/1173. 

It cannot be said that San'a 3 was constantly in the 
hands of the Ayyubids during their period in the 
Yemen. Rather, the Ayyubids fought from time to 
time against those in control in the town, firstly the 

Hamdanid sultans, later the Zaydis, and always after 
fairly elaborate military manoeuvres to secure 
Dbamar to the south of San c a 3 . They took over the 
town for a period, until ousted, and then the whole 
process began all over again after an interval of time. 

A similar story can be told of the Rasulid period 
during the years 628-723/1230-1323. Despite the 
brilliance of Rasulid administrative and intellectual 
achievement in Tihama and the southern highlands, 
they too never succeeded in occupying with anything 
like permanency the chief town of the country. 
During approximately the last 130 years of the dynas- 
ty, 723-858/1323-1454, San<a> remained in general 
within the political orbit of the Zaydis and beyond the 
grasp of the Rasulids and indeed of their successors, 
the Tahirids [qv.] (see Smith, Some observations on the 
Tahirids and their activities in and around SanV (858- 
92311454-1517), in Ihsan Abbas a alii (eds.), Studies in 
history and literature in honour of Nicola A. Ziadeh ..., Lon- 
don 1992, 29-37). (See Smith, SanV, 49-68). 
Late mediaeval and modern SanV 

Again, elaborate details are not possible here and 
the following is a very brief outline of dates and 
political events. From about 954/1547, when the Ot- 
toman Ozdemir Pasha [q. v. ] advanced on San'a 3 , un- 
til 1038/1629 San c a 5 was the capital of Ottoman 
Yemen. Although he did not deal the coup de grace — 
that was left to the Imam al-Mu'ayyad— the real hero 
of the Turks' expulsion from San'a 3 and the Yemen 
was al-Kasim b. Muhammad, al-Kasim al-Kablr. 
Right through to the mid-13th/19th century, the 
Zaydl Imams controlled San'a 3 and northern Yemen 
with more or less authority. ZaydT rule, however, 
seriously declined, mainly as a result of internal 
squabbling, allowing a second Ottoman occupation of 
the Yemen. The Ottomans had already begun to ex- 
pand in Arabia and in 1289/1872 Ahmad Mukhtar 
Pasha entered San'a 3 to begin the second Ottoman oc- 
cupation of the Yemen until about 1337/1918. The 
post-1337/1918 history of San c a> and the Yemen is 
that of the Zaydl Hamld al-DTn house, under the 
Imams Yahya, Ahmad and, for a few days only in 
1382/1962, al-Badr. The house rose to prominence in 
Zaydl circles in 1307/1890 in the figure of al-Mansur 
bi 'llah Muhammad b. Yahya Hamld al-Dln. The 
Imam Ahmad (1367-82/1948-62), famed for his 
bravery and learning as much for his ruthlessness and 
toughness, died from natural causes in 1382/ 
September 1962. His son, al-Badr, was proclaimed 
Imam and widely recognised, but was compelled to 
flee San'a 3 after he was attacked by young army of- 
ficers who had been plotting a military coup. Hamld 
al-DTn rule was brought to an end and the newly- 
appointed chief of staff , c Abd Allah al-Sallal, was pro- 
nounced the new president of the Yemen Arab 

The Republic needed vast numbers of Egyptians 
troops to prop it up in the face of general royalist op- 
position. They left the Yemen only in 1967 at the time 
of the Six-Day War between Egypt and Israel and the 
Republican government, balanced between the Zaydl 
and Shaf^i religious and tribal interests, the military 
and a new breed of Western-educated technocrat, 
continued in power and opened up talks with the 
People's Democratic Republic of Yemen on the ques- 
tion of the formation of a unified Yemen. The 
Republic of Yemen came into being in 1990 with 
San^ as its capital (see Serjeant, in SanW, 68-108). 
The buildings of SanW 

The visitor to the old town of San c a 3 (which is still 
largely intact though surrounded by new develop- 
ment) notices the extremely prominent traditional 

domestic architecture, the high, multi-storey tower 
house. It is usually square and at least five, if not eight 
or nine storeys high. The ground floor is used as stores 
and for the domestic animals and the top storey, called 
by the name mafradj., is used as a second reception 
room and for the daily afternoon ia(-chewing ritual 
(see Lewcock and Serjeant, in SanV, 436-500). 

Another extremely prominent architectural feature 
in San'a' is naturally the mosque and there are over 
one hundred in the town. Al-HadjarT (Masadjid SanV, 
San'a 3 1361) lists about a hundred of many different 
periods and is a mine of information regarding their 
history and general background. Perhaps the most 
impressive and interesting for the historian is the 
Great Mosque (al-Diami^ al-Kabir) whose original 
foundation no doubt has an early Islamic date (al- 
Mad'adj, The founding of the Great Mosque ..., in New 
Arabian Studies, i [1993], 175-88, suggests the first con- 
struction was in 11/633, p. 184). For a detailed study 
of the mosques of San'-a 3 , see Serjeant, Lewcock, 
Smith and Costa, in SanW, 310-90). 

The public baths (hammdmdt) play an important 
social role in the daily life of San'a 3 and there are 
seventeen still in operation today, some possibly pre- 
Islamic in foundation. The fuel burnt is human excre- 
ment (khara), which is reputed to burn giving off great 
heat. The fuel is collected from the "long drop" (man- 
tal) found in every house in the town for this very pur- 
pose. Women have their own bath times regulated at 
each public bath and the baths play a particularly im- 
portant role at festival times (a'-yad) (see Lewcock, 
Akwa c and Serjeant, SanW, 501-25). 

The market of SanV 

The large market in San'a 3 has been the subject of 
some academic studies (see in particular Serjeant, 
Akwa c and Dostal, in SanV, 159-302) and the com- 
plexities of its administration and organisation are be- 
ginning to be understood. A 12th/18th century docu- 
ment entitled Kanun San'-P, which is a collection of 
market regulations, was published by Husayn b. 
Ahmad al-Sayaghl in 1964 (Madjallat al-Makhtutat, 
273-307) and it has also been the subject of a lengthy 
study in SanV, 179-240. Today's San'a 3 market is 
dealt with in some detail by Dostal, in SanV, 241-75. 
The mint of SanV 

The earliest coins which can be assigned to San'a 3 
date from 156-8/772-4, and the 3rd/9th century in 
particular saw a huge output from the mint, "a 
substantial proportion of all the gold being coined in 
the territories of the caliph", according to Lowick (in 

Bibliography: It is difficult to exaggerate the 
scholarly importance of R.B. Serjeant and R. Lew- 
cock (eds.), SanW, an Arabian Islamic city, London, 
1983, written by a whole team of Yemen specialists, 
for almost any aspect of the history, culture and dai- 
ly life of the town. Other sources not mentioned in 
the text are: Husayn c Abd Allah al- c Amri and 
Yusuf Muhammad c Abd Allah, art. San c d>, in 
Ahmad Djabir c Afif et alii (eds.), al-Mawsu c a al- 
Yamaniyya (The Encyclopaedia of Yemen), San c a> 1992, 
ii, 583-8. History: Ahmad b. c Abd Allah al-RazI, 
Tarlkh madinat SanW, ed. Husayn b. c Abd Allah al- 
c Amn and c Abd al-Djabbar Zakkar, 2 Damascus 
1981; A.S. Tritton, Rise of the Imams of Sanaa, 
Oxford-Madras 1925; J.R. Blackburn, The collapse 
of Ottoman authority in Yemen, 968/1560-976/1568, in 
WI, xix (1979). Buildings: P. Costa, La Moschea 
grande di SanW, in AIUON, xxxiv (1974). Mint: 
RamziJ. Bikhazi, Coins of al-Yaman, 132-569 AH, 
in al-Abhdth, xxxiii (1970). (G.R. Smith) 

SANAD [see isnad]. 

SANADJAT, weights of a balance (in full 
sanadjat al-mizan); also applied to balances, steelyards; 
also the weights of a clock (sing, sandja). The forms 
with sad also occur (sanadjat and sandja) but the former 
is the more chaste (see Lane, s.v.). There are two 
recognised plural forms, sanadjat and sinadj (in modern 
Egyptian Arabic sinag, plural of singa). The word is 
Persian in origin, being connected with sang, meaning 

were non-metallic (cf. 'the Hebrew of Deut. xxv, 13). 
According to Muslim tradition, it was a Jew named 
Sumayr, during the time of al-Hadjdjadj b. Yusuf 
[?.».], who first proposed to regulate the new dirhams 
of the reformed coinage of 75/694 by means of the use 
of fixed weights (Ibn al-Athlr, iv, 337). Previously, 
the custom apparently had been to weigh one coin of 
good quality against another. When a large number 
had thus been weighed, this lot was weighed against 
a similar number and the surplus, if any, was carried 
forward. The first coin weights of Islam were made of 
bronze and are excessively rare. Weights of iron are 
also recorded but no examples are extant. Under the 
Umayyad caliph c Abd al-Malik (65-86/685-705 
[q.v. ]), weights made of glass were recommended to 
be used since they did not change by increase or 
decrease (al-Damlri, Haydt al-hayawdn, i, 59). This 
carried on the practice of Ptolemaic and Byzantine 
times. These glass weights, however, were confined to 
Egypt and were in use from Umayyad until Mamluk 
times. The old opinion that they were glass coins, 
nummi vitrei, was first exploded by Castiglioni in 1847, 
and later, after the fact had been overlooked, by E.T. 
Rogers in 1873. Various collections of these sanadjdt 
have been published. As they generally bear inscrip- 
tions with the names of caliphs, or governors, or in- 
spectors of markets, and an indication of weight, they 
are very valuable, not only for Islamic history and 
metrology, but also for Arabic epigraphy. 

The great physicist al-Khazinl [q. v. ] paid particular 
attention to the weights used in his balances [see 
mizan] , in which of course he achieved a very high 
degree of accuracy. In his introduction to his descrip- 
tion of the Balance of wisdom, he devotes two 
paragraphs to a discussion of the weights specific to 
this machine. He does not seem to imply that the 
sanadjat — always with sad in his work — can be in 
mixed units in weighing operations, but simply says 
that they are to be "dirhams, milhkdh or others". 

Presumably the units chosen in any particular opera- 
tion depended upon the mass of the objects to be 
weighed. For each unit there were nine different 
weights of sanadjat: for unity, 1, 2 and 5; for tens, 10, 
20 and 50; for hundreds, 100, 200 and 500. With 
these values, using different combinations, any re- 
quired weight could be obtained (Kitab Mizan al- 
hikma, Haydarabad 1940, 108-9). 

The term sanadjat is also occasionally used for other 
kinds of weights such as counterweights or pellets 
discharged from the mouths of falcons in water-clocks 
(see E. Wiedmann, Aufsdtze zur arabischen Wissen- 
schaftsgeschichte, Hildesheim 1970, i, 127). More usual 
terms for weights where the precise value is unimpor- 

Bibliography: Lane, s.v.; Dozy, Suppl. , s.v.; 
CO. Castiglioni, Dell'uso cui erano i oetri con epigrafi 
cufiche, Milan 1847; E.T. Rogers, Glass as a material 
for standard coin weights, in Num. Chron. (1873), 60- 
88; idem, Unpublished glass weights and measures, in 
JRAS (1872), 98-112; S. Lane-Poole, Arabic glass 
coins, mNum. Chron. (1872), 199-211; idem, Cat. of 
Arabic glass weights in the British Museum, London 
1891 ; P. Casanova, Etude sur les inscriptions arabes des 
poids et mesures en verre, Cairo 1891 ; idem, Deneraux en 
verre arabes, in Melanges offerts a M. Gustave 
Schlumberger, Paris 1924, 296-300; idem, Cat. des 
pieces de verre des epoques byzantine et arabe de la Collection 
Fouquet, in MMAF(1893), 337-414; J. B. Nies, Kufic 
glass weights and bottle stamps, in Procs. of American 
Num. and Arch. Soc. (New York 1902), 484-55; Sir 
W.M.F. Petrie, Glass weights, in Num. Chron. 
(1918), 111-16; idem, Glass stamps and weights 
(University College of London Collection), London 
1926; A. Grohmann, Arab. Eichungsstempel, Glass- 
gewichte und Amulette aus Wiener Sammlungen, in 
Islamica (1925), 145-226; R. Vasmer, art. Sandj in 
F. von Schrotter's Wbrterbuch der Munzkunde, Berlin 
1930; H. Sauvaire, Materiaux pour 

la ; 

, Paris 

1882, 19 ff.; Lavoix, Cat. des monn 
Paris 1887, i,'xlv-xlvi; E. von Zambaur, in Num. 
Zeitschrift (1903), 317-19; M. Jungfleisch, Un Poids 
fatimite en plomb, in BIE (1926-7), 115-28; idem, 
Poids fatimites en verre polychrome, in ibid., 19-31; 
idem, Les rolls discoides en verre, in ibid. (1927-8), 61- 
7 1 ; J . Farrugia de Candia, Deneraux en verre arabes, in 
RT (1935), 165-70; W. Airy, On the Arabic glass 
weights, London 1920; Rev. Numismatique (1906), 
225; Damni, Haydt al-hayawdn, i, 80 (the English tr. 
by Jayakar (i, 128) wrongly translates sanadjat as 
"scales"); J. Walker, in Num. Chron. (1935), 246-8. 

(J. Walker-[D.R. Hill]) 
SANA'I, Madjdud b. Adam al-Ghaznawi, Per- 
sian poet. In early sources already the kunya Abu '1- 
Madjd is sometimes added to his name. As a pen 
name he used Sana 5 !, only rarely Madjdud or 
Madjdud Sana 5 !. The former name could have been 
derived from Sana 3 al-Milla, one of the lakabs of the 
Ghaznawid sultan Mas'ud III, but the poet's actual 
relationship to this ruler is unclear, because no 
panegyrics directly addressed to him by Sana 3 ! have 
been preserved. As a matter of fact, no reliable 
biographical data outside the poet's own works are 
available. However, the many references to the 
historical context to be found in his poetry, including 
dedications to a great number of patrons, make it 
possible to reconstruct his life, at least in its main 

The only known dating of his birth, in 437/1045-6, 
which is recorded in the Muajmal-i Fasihi, is highly im- 
probable. It may be regarded as certain that he was 

born at Ghazna about the beginning of the last 
quarter of the 5th/l 1th century. In spite of the reputa- 
tion of a learned poet, expressed by the epithet hakim 
commonly added to his name, we know nothing about 
his formal education beyond the fact that in a prose 
introduction to the Hadikat al-hakika his father is said 
to have been a mu'-allim. According to the picture of 
his early life sketched in the Karnama-yi Balkhi and the 
panegyrics contained in his Diwdn, he was during 
these years a minor poet who maintained connections 
with patrons from all social groups in the Ghaznawid 
residence except, as it seems, the court of the sultan 
itself. We find among them officials of the state 
bureaucracy, military men, members of the Islamic 
clergy as well as scholars, scribes and poets. One of his 
most important protectors was Thikat al-Mulk Tahir 
b. C A1I, the head of the department of correspondence, 
who also patronised the poet Mas c ud-i Sa c d-i Salman 
[q.v.]. According to a kit'-a in Sanaa's Diwan, he 
prepared a collection of Mas c ud's poetry. There is 
also mention of his copying the poems of another 
prominent poet, c Uthman Mukhtarf [q.v.]. 

His relations with a few prominent Islamic 
scholars, belonging to the Hanafi school of law, 
foreshadowed the future course of his literary career. 
The celebrated story of Sanaa's abandoning profes- 
sional poetry after a meeting with a "drinker of 
dregs" (lay-kjfdr), first told by Dawlatshah towards 
the end of the 9th/15th century, has no historical 
value, but does reflect an evident break in his career 
as a poet, more or less coinciding with his departure 
from Ghazna. First he went to Balkh, and from there 
to other cities in Khurasan, which was then under the 
rule of the Saldjuks. As it appears from the Karnama-yi 
Balkhi, written shortly after he left Ghazna, this event 
took place before 508/1114-15, the year of Mas c ud 
Ill's death. He probably did not return to his native 
city before about 520/1126. During this exile, he al- 
most completely abandoned the secular poetry of the 
court tradition, seeking instead the patronage of 
people belonging to the religious class. Of crucial im- 
portance was his stay at Sarakhs. where he established 
a close relationship with Sayf al-Hakk Abu '1- 
Mafakhir Muhammad b. Mansur. This Hanafi 
scholar was, as his title akdd 'l-kuddt indicates, also a 
person of some political weight. Besides, he was a 
renowned preacher {waHz), who delivered his sermons 
in a khdnkdh of his own. Sana 3 ! wrote some of his most 
remarkable poems for Muhammad b. Mansur, ad- 
dressing him both as a protector and as a spiritual 
guide. There are further references to the poet's visits 
to Nishapur and Harat. In the latter town, he met 
with the descendants of Shaykh c Abd Allah al-AnsarT 
[q.v.]. This is in fact the only contact of his with a Sufi 
community for which there exists reliable evidence. 
The tradition that Sana 3 ! was a pupil of Shaykh Yusuf 
al-Hamadhani. which eventually led to his inclusion 
in the Sufi affiliations (cf. R. Gramlich, Die schiitischen 
Derwischorden Persiens, i, Wiesbaden 1965, 8), is cer- 
tainly unhistorical. 

The reasons for his return to Ghazna are not known 
to us. However, again a remarkable change in his 
career can be noticed. While he was still in Khurasan, 
Sana'i had become a celebrated writer of religious 
poetry for which there appeared to exist a lively in- 
terest among the ruling classes, who hitherto had only 
patronised secular poetry. For the first time in his 
career, Sana 3 I drew the attention of a royal patron 
when the Ghaznawid sultan Bahramshah [q.v.] in- 
vited him to the court. Although he clearly stated his 
determination to stay aloof from the affairs of this 
world, he nevertheless wrote his major mathnawi as 

well as a number of short poems for this sultan. Very 
different dates are suggested in later sources for the 
year of his death. The most likely is 11 Sha c ban 525/9 
July 1 1 3 1 , mentioned in a notice on the last day of his 
life, which became attached as an appendix to a prose 
introduction to the Hadika. The same date is men- 
tioned in the Nafahdt al-uns of Djami [q. v. ] . 

The works of Sana 3 ! encompass the entire range of 

classical Persi 
with religious 
poetry are by 
poems often coi 

poetry. Though most poems deal 
lbjects, specimens of purely secular 
o means absent. Even the religious 
ain panegyrical elements showing the 
poet's dependence on the material support of patrons 
throughout his career. The Diwan of Sana 3 ! is pre- 
served in a number of ancient manuscripts, the oldest 
dated of which is Velieddin 2627 (Bayezit Library, 
Istanbul), copied in 684/1285. They show great dif- 
ferences, as far as the order of the poems, variant 
readings and the number of verses are concerned. It 
is, therefore, impossible to establish which of them 
could be taken to represent a genuine textual tradition 
going back to the time of the poet himself. In one 
strain of the early tradition, the poems were grouped 
into a few sections marked by generic indications. The 
most important genres thus distinguished are the 
zuhdiyyat, ' 'ascetic poems" (most of them long kasidas) 
and the kalandariyydt, poems characterised by the use 
of antinomian motives referring to the debauchery of 
beggars and drunks. Of considerable interest is also 
the group of the ghazaliyyat because this is the earliest 
sizable corpus of this kind of poems known in the 
history of Persian literature. 

As a writer of mathnawis, Sana 3 i is best known for 
his long didactical poem which is usually entitled 
Hadikat al-hakika wa-sharfat al-larika, "The garden of 
truth and the law of the right path". The history of 
this text is even more complicated than that of the 
Diwan. To all appearances, the poet died before he 
could give the poem its final form. At least one early 
version, however, was prepared for Sultan Bahram- 
shah, when the poet was still alive. A copy of this ver- 
sion, under the title Fakhri-nama, has survived in the 
manuscript Bagdath Vehbi 1672 (Siileymaniye 
Library, Istanbul), which was copied at Konya on 7 
Shawwal 552/12 November 1157. It contains about 
5000 bayts, i.e. approximately half the amount of lines 
found in most later copies. It is not likely, however, 
that all the lines not contained in the Vehbi 
manuscript are unauthentic. The textual tradition has 
preserved a few documents which indicate that inten- 
sive editorial work was done on the poem soon after 
the first version was completed, including a rough 
copy (musawwada) of 10,000 bayts prepared by the poet 
to be sent to Kh w adja Burhan al-Din, a scholar from 
Ghazna who lived at Baghdad, after the poem had 
come under attack for its alleged pro-'Alid tendencies. 
After Sana 3 T's death, Bahramshah ordered a certain 
Muhammad b. C A1T al-Raffa 3 to make yet another 
redaction. As a result of these initial rearrangements, 
as well as of subsequent editorial interference, the 
Hadika was transmitted in several different forms. As 
late as the llth/17th century ,the Indian scholar c Abd 
al-Latif al-'Abbasi made an attempt to harmonise the 
various traditions of the text; he also wrote a commen- 
tary, entitled Latd'if al-haddHk min nafdHs al-daRaHk. 
Later a few other commentaries were composed (cf. J . 
Stephenson, The enclosed garden of the truth, repr. Well- 
ingborough 1975, Introd., pp. xxi-xxv). One of the 
selections made from the poem, which is sometimes 
entitled Lalifal al-Hrfan, has been ascribed to Sana'I as 
well as to Farid al-Din c Attar [q. v.], but its real author 
is probably Nizam al-Din Mahmud Husayni ShirazI, 

a poet of the 9th/ 15th century also known by the pen 
name of Da c T. 

The Hadikdt al-hakika is the first specimen of a 
mystical mathnawi in Persian literature and has had a 
considerable impact on later writers in the same 
genre, notably on Djalal al-DIn Ruml [q.v.], whose 
Mathnawi-yi maHiawi was composed after the example 
given by Sana 3 !. The great number of manuscripts 
known to exist bears witness to an immense populari- 
ty lasting throughout the centuries. It is a didactic 
poem conceived as a continuing discourse on a wide 
variety of ethical and religious subjects. In the early 
Fakhri-ndma version, the vague outline of an allegory, 
comparable to that of his earlier work Sayr al-Hbad ila 
'l-ma'-ad, appears when the poet tells of a meeting with 
a spiritual guide, who is the personification of the Ac- 
tive Intellect. In later redactions this feature was al- 
most obliterated and replaced by a division of the text 
into chapters. The text contains numerous references 
to philosophy and the sciences and has therefore often 
been called an "encyclopaedia of Sufism". This is a 
misleading qualification because these elements are 
always subordinated to the didactic discourse. The 
same applies to the narratives, which take a far less 
important part in Sana'I's poem than they do in the 
works of later writers of Persian didactic mathnawis. 

Sana'i left two other mathnawis, of a much smaller 
size. The first, Kdrndma-yi Balkhi, sometimes also 
called Mutdyaba-ndma ("Book of jest"), is a completely 
secular poem of no more than 433 bayts. It was written 
at Balkh shortly after Sana 3 ! had left Ghazna. In a 
mixture of praise and satire, this topical poem reviews 
the people who were the poet's patrons during his ear- 
ly years, arranged according to their social position. 
The second, the Sayr al-Hbad ila 'l-ma'-ad ("The 
journey of the devotees to the place of return") is 
Sanaa's most interesting work. Two-thirds of its 800 
lines describe the development of the narrator's soul 
in the allegory of a spiritual journey. From his concep- 
tion onwards, he climbs the ladder of existence, partly 
under the guidance of the Active Intellect. He reaches 
the goal of this quest when he meets with the preacher 
Muhammad b. Mansur, his actual patron during his 
stay at the city of Sarakhs. whose praise fills the re- 
maining part of the poem. The Sayr al-Hbad is written 
in an enigmatic style with only few explanations pro- 
vided in the text. Stylistically it shows a resemblance 
to philosophical allegories in Arabic, such as Ibn 
Slna's Hayy ibn Yakzdn [q.v.] and the Lughz kabis by 
Miskawayh [q. v.]. An anonymous commentary is ex- 
tant, the oldest version of which is contained in the 
ms. Nafiz Pasa 410 (Siileymaniye Library, Istanbul), 
dated 674/1275. 

All genuine mathnawis of Sana'i were written in the 
metre khafif, which, through his example, became one 
of the patterns most often used for didactic mystical 
poetry. In the same metre, a number of other short 
mathnawis occur in manuscripts of his works, which 
are falsely attributed to him. Some of these were 
works by other poets which were manipulated in order 
to pass them off as genuine texts by Sana'I, e.g. 'Ishk- 
ndma, which actually is a verse commentary on the 
Sawdnih of Ahmad al-Ghazall [q.v.], and Tarik al- 
tahkik, written in imitation of the Hadika by Ahmad al- 
Nakhciwanl who probably lived in the 8th/ 14th cen- 
tury (on these and other poems of this group, see B. 
Utas, Tariq ut-tahqiq, Lund 1973, passim). Others, like 
Tahrimat al-kalam and 'Akl-ndma, both appearing for 
the first time in the Velieddin manuscript of 684 
A.H., were deliberately composed as pseudo-Sana'! 
texts (cf. Of piety and poetry, 113-18). 

A small collection of letters by Sana°i was published 

- SANAM 5 

by Nadhlr Ahmad (Makdtib-i Sana^i, Aligarh 1962). 
There are two prose introductions to his works con- 
taining his name as the author. One of these, which 
is in fact the same text as the introduction by al- 
Raffa 3 , is almost certainly spurious. 

Sana'I's great impact on Persian mystical poetry 
has given rise to the view that he was a prominent Sufi 
himself. There is little historical evidence available to 
substantiate this view. The most important patrons of 
his art were Islamic scholars, many of whom were also 
renowned as preachers. The predominently homiletic 
style of his religious poetry fits this social environment 
extremely well. With its characteristic blend of ethics, 
wisdom, mysticism and praise of the Prophet and 
other great men of Islam it appealed to the community 
of the Muslims as a whole rather then to a restricted 
circle of Sufi adepts only. This also explains the varied 
use made by later generations of his verse. 

Bibliography: An extensive bibliography can be 
found in J.T.P. de Bruijn, Of piety and poetry: the in- 
teraction of religion and literature in the life and works of 
Hakim Sand^i of Ghazna, Leiden 1983. To be added 
are: idem, The transmission of early Persian ghazals, in 
Manuscripts of the Middle East, iii (1988), 27-31; 
idem, The Qalandariyydt in Persian mystical poetry, in L. 
Lewisohn (ed.), The legacy of mediaeval Persian Sufism, 
London-New York 1992, 75-86; idem, Comparative 
notes on Sand^i and '■Attar, in Lewisohn (ed.), Classical 
Persian Sufism: from its origins to Rumi, London-New 
York 1993, 361-79; idem, The stories of Sandy's 
Faxri-ndme, in Melanges offerts a Charles-Henri de 
Fouchecour, ed. by Ziva Vesely (forthcoming); 
Sana-T, Viaggio nel regno del ritorno, Parma 1993, con- 
tains an Italian translation of the Sayr al-Hbad ila '1- 
ma'dd, with an introduction and notes, by Carlo 
Saccone; M.R.Sh. Kadkanl, Tdziydnahd-yi suluk. 
Nakd wa tahlil-i (and kasida az Hakim Sand% Tehran 
1372/1993. Q.T.P. de Bruijn) 

SANAM (a.), image, representation and, 
especially, idol (from the Common Semitic root s-l-m, 
cf. Akk. salmu, Aram, salmd, Hebr. selem, etc., by a 
shift of / into n, see Gesenius-Buhl, 684); for Old 
Testament parallels, see inter alia, Num. xxxiii. 52; II 
Kings xi. 18; Ezek., vii. 20; Amos, v. 26). It is in this 
sense that it is found in the Kur'an, where the pi. 
asndm is cited five times (VI, 74; VII, 138; XIV, 35; 
XXI, 57; XXVI, 71). 

Sanam progressively replaces nusub (pi. ansdb, Hebr. 
masfbot, Gen. xxxv. 14), a term denoting "carved 
stones over which the blood of victims sacrificed to 
idols was poured, stones making up tumuli and those 
delimiting the sacred enclosure (hima) of the sanc- 
tuary" (T. Fahd. Pantheon, 26). From being the rough 
stone making up the nusub, the idol became "a carved 
stone" (Yakut, iv, 622; fa-nahatahu 'aid swat asndm al- 
Bayt [i.e. the Ka'ba]). Ibn Hisham writes (Sira, 
54 = al-Azrakl, Akhbdr Makka, 78) that all the in- 
habitants had an idol (sanam) which they worshipped, 
and whenever one of them left on a journey, the last 
thing he did before leaving and the first on his return, 
was to touch the idol (iamassaha bihi; cf. Gen. xxxi. 
14). Hence, Yakut says, the cult of stones amongst the 
Arabs in their encampments began from their deep at- 
tachment to the idols of the Haram (loc. cit.). This cult 
reflects a state of affairs well before the reform of c Amr 
b. Luhayy in the 3rd century A.D., which introduced 
into central Arabia statues of gods brought in from 
Hit (Heliopolis = 'Amman), where he had sought 
medical treatment. "In effect, the teraphim of the Ca- 
naanites, the eldhim of the Hebrews and the Hani of the 
Assyrians long survived up to the time of the coming 
of monotheism in the shape of betyles in stone, in sand 


held together with milk, in flour and in wood, in 
pagan Arabia" (see refs. in Fahd, op. cit. , 28-9). "It 
was in the mid-3rd century A.D. that Nabataean and 
Syro-Palestinian influences affected, in the urban cen- 
tres, the evolution in form of the Arab pantheon, and 
it was then that the stone betyle became an idol". 
Wellhausen correctly affirmed this when he wrote that 
the images were not genuinely Arabic, since wathan 
and sanam were imported words and things. See his 
Reste 7 , 102; D. Nielsen, Die altarabische Mondreligion und 
die mosaische Uberlieferung, Strassburg 1904). 

According to al-Azraki, there was no house in Mec- 
ca without its idol. Those carved out of wood seem to 
have been those of the high-born and rich, as were im- 
ported products. Ibn Hisham states (Sira, 303-4) that 
" c Amr b. Djamuh, who was one of the sayyids of the 
Banu Salama and one of their leaders, had taken into 
his house a wooden idol (sanam), called Manat, ac- 
cording to the practice of notables of that time; he 
considered it as a god, and worshipped and purified 
it". c Ikrima, son of Abu Djahl [q.v.], the fierce enemy 
of the Prophet, made idols. Traders went and offered 
them to the Bedouins, who bought them and placed 
them within their tents (al-Azraki, 77-8). 

After the triumph of Christianity in the Orient, the 
Hidjaz had remained a stronghold of paganism, 
where idol carvers could still gain a living. It was not 
surprising that, when the Prophet entered Mecca in 
triumph, he had 360 idols in the Ka c ba immediately 
destroyed (ibid, 77; Ibn al-Athlr, ii, 192; cf. Fahd, 
Pantheon, 31). According to a leading Jordanian 
historian, met by the author in Kuwait shortly after 
the expulsion from the Ka c ba in 1979 by French 
guards of the rebels who had barricaded themselves 
there, the ground collapsed as the result of a bomb ex- 
plosion and revealed a pile of statues buried below 
ground; the Saudi authorities hastened to dispose of 


;an Arabia, froi 

Thus the relij 
carved idols, retained their pr 
ture, and although social conditions and artistic influ- 
ences might affect the actual forms of representation, 
this had no effect on the conceptual development of 
the cults there. A list of the Arab gods can be found 
in Fahd's Pantheon, with all the known information 
about some 90 of them, and showing that these Arab 
cults were static and characteristic of the desert and 
nomadic life, in which betyles, idols and sacred trees 
were the dominant features. 

Bibliography: This article is based on T. Fahd's 

Le pantheon de l' Arabic Centrale a la veille de I'hegire, 

Paris 1968; see also idem, in Mythes et croyances du 

monde entier, ii, Paris 1985, section "Le pantheon 

arabe avant lTslam". Amongst references used in 

the above, still of value, are: Wellhausen, Reste 1 , 

Berlin 1897; E. Dhorme, Religion primitive des 

Semites, in RHR, cxxviii (1944), 1-27; idem, Les 

religions arabes preislamiques , review of G. Ryckmans, 

in ibid., cxxxiii (1947), 34-48; G. Ryckmans, Les 

religions arabes preislamiques 2 , Louvain 1953 ( = Bibl. 

du Museon, xxvi = A. Quillet, Hist. gen. des religions, 

Paris 1960, ii, 199-228); Djurdji Zaydan, Ansdb al- 

'■Arab al-kudama*, Cairo 1906. See also hubal; isaf 

wa-na'ila; kaws kuzah; al-lat; manaf; rabb; sa c d 

wa-nahs; shams. (T. Fahd) 

SANAM KADIS, see kadis, at vol. IV, 583-4, and 

add to the Bibl. there: Abu Hamid al-Gharnatl, ed. 

Ferrand, the main reference, now tr. Ana Ramos, Abu 

Hamid al-Garnati (m. 56511169), Tuhfat al-albab (El 

Regalo de los espiritus), Madrid CSIC-ICMA 1990, 

(fuentes Arabico-Hispanas 10), 46-7, for the passage in 

question ( = ed. Ferrand 69-70), with representation 

here of the plate ( = Ferrand, between 200 and 201) 
showing the "object", i.e. the lighthouse (which 
Ramos renders by "statue", estatua). See also P. Mar- 
tinez Montavez, Perfil de Cadiz hispano-drabe, Cadiz 
1974; Sahar al-Sayyid c Abd al- c AzTz Salih, Madlnat 
Kadis wa-dawruhd fi 'l-ta^rikh al-siydsi wa 'l-hadari li 'l- 
Andalus fi 7-W al-islami, Alexandria 1990; D. 
Bramon, El mundo en el sigh XII. El tratado de al-Zuhri, 
Sabadell 1991 ( = Span. tr. of the K. al-Dj.a c rqfiyya, ed. 
M. Hadj-Sadok). (J. -P. Molenat) 

SANANDADJ or sinandadj, older form sinna, the 
administrative capital of the modern Persian pro- 
vince of Kurdistan and the general name for the 

1 .The to wn . The name Sinna came into historical 
prominence only from the 9th/15th century onwards, 
the main urban centre of the district having preciously 
been Sisar [q.v.], as the seat of the Kurdish waits or 
local rulers of Ardalan [q.v.]. Under the year 
988/1580, the 10th/ 16th century historian of the 
Kurds, Sharaf al-DIn Khan Bidllsl [q.v.], speaks in his 
Sharaf-nama (ed. V. Veliaminof-Zernof, St. Peters- 
burg 1860-2, i, 88) of a land grant held by Tfmur 
Khan of Ardalan, which included Hasanabad, Sina, 
etc., but the local historian of Sanandadj, C A1I Akbar 
Munshl Wakayi c -nigar (wrote ca. 1310/1892-3, see 
Storey, i, 1300, cf. also 369), in his ms. work on the 
geography and history of Sanandadj, the Hadika-yi 
Nasirl (Fr. resume by B. Nikitine, in RMM, xlix, 70- 
104), says that Sinna was built by the wd/f Sulayman 
Khan on the ruined site of an earlier settlement; the 
Persian ta?nkh or chronogram for this event is given as 
ghamha "sorrows" = 1046 in abajad [/1636-7]. 

The walis, being near the Perso-Turkish frontier, 
took an active part in the wars between the Safawids 
and the Ottomans, sometimes on one side, sometimes 
on the other; during this period, Ardalan enjoyed, like 
c Arabistan and Luristan, a semi-independent ex- 
istence. During the Afghan invasion of Persia, in 
1132/1720, the Kurdish chief of Sulaymaniyya [q.v.] 
seized Sinna, but the wall of Ardalan Subhan Werdi 
Khan recovered it in the time of Nadir Shah. Aman 
Allah "the Great" (1212-40/1797-1825) much im- 
proved the town of Sinna, and entertained there Sir 
John Malcolm and J.C. Rich, but his grandson Aman 
Allah b. Khusraw Khan Na-kam (1265-84/1848-67) 
was the last hereditary wall of Kurdistan, since the 
central government in Tehran under Nasir al-Din 
Shah [q.v.] now extended its direct control over the 
western provinces. 

The town lies between the right bank of the Kishlak 
river (which eventually flows jlown to the Diyala and 
then the Tigris) and Mount Awidar, which separates 
Sinna from the old capital Hasanabad. The citadel of 
the waits, described by such 19th century travellers as 
Malcolm, Rich and Cirikov, is on a hill some 21 m/70 
ft. high in the centre of the town. The town's popula- 
tion at that time was largely Kurdish, but with a 
substantial Jewish minority and a few Chaldaean 
Catholic and Armenian Christians, and it was a lively 
commercial centre, exporting oak-galls, tragacanth, 
furs and carpets. In modern Persia, Sanandadj (lat. 
35° 19' N., long 47° 01' E.) is the administrative 
capital of the province (ustan) of Kurdistan, with a 
population in 1991 of 244,249 (Preliminary results of the 
1991 census, Statistical Centre of Iran, Population 
Division). Senna carpets remain a principal product 
of the town and its hinterland. 

2. The district. The older district of Sinna cor- 
responds with the heartland of the modern Kurdistan 
province, with high, treeless plateaux in the northeast 
and southeast, whilst its central part is intersected by 


numerous valleys, sloping down to deciduous forests 
in the east. The main mountain massif is the Kuh-i 
Cihil Cashma (ca. 3,658 m/12,000 ft.) to the north of 
Sanandadj and running westwards to the c IrakT fron- 
tier. For a detailed survey of topography, see Minor- 
sky's EI 1 art. Senna. 

Ethnically, the population is mainly Kurdish, Sun- 
nis of the Shafi c I madhhab, amongst whom shaykhs of 
the Nakshbandiyya [ q. v. ] Sufi order have traditionally 
been influential, although the walls of Ardalan were 
Shi*-!, possibly as a result of former connections with 
the Ahl-i Hakk [q.v.] Guran. In addition to the ma- 
jority of Kurdish speakers, the area to the west of 
Sanandadj, going up to and slightly across the frontier 
with c Irak, contains speakers of the Hawramam or 
Awromani dialect of the Northwest Iranian tongue 
GuranI [see guran; hawraman; and to the references 
there, add D.N. MacKenzie, The dialect of Awroman 
(Hawramdni-i Luhdn), in Royal Danish Acad, of Sciences, 
phil.-hist. series, iv/3 (Copenhagen 1966), pp. 141]. 
Bibliography: Sharaf al-Din BidlTsI, Sharaf -nama, 
i, 82-9, 317-19, 320-22; J. M. Kinneir, A geographical 
memoir of the Persian empire, London 1813, 142-7; Sir 
R. Ker Porter, Travels in Georgia, Persia, Armenia, 
Ancient Babylonia, etc., London 1822, ii, 540-55, 
563-8; Sir John Malcolm, History of Persia, Fr. tr. 
Paris 1821, iii, 302; idem, Sketches of Persia, London 
1827, ii, 273; J.C. Rich, Narrative of a residence in 
Koordistan, etc., London 1836, i, 185-248; 281; Rit- 
ter, Erdkunde, ix, 412-59; Chevalier T.M. Lycklama 
a Nijeholt, Voyage en Russie, au Caucase et en Perse, 
Paris-Amsterdam 1872-5, iv, 30-70; G. Hoffmann, 
Ausziige aus den syrischen Akten der Martyren, Leipzig 
1880, 265-6; J. de Morgan, Mission scientifique en 
Perse. Etudes geographiques, Paris 1895, ii, 47-61; M. 
Streck, Das Gebiet der heutigen Landschaften Armenien, 
Kurdistan und Westpersien, in ZA, xiv, 138-9; Ad- 
miralty Handbooks, Naval Intelligence Division, 
Persia, London 1945, 54, 303, 468-70. 

_(V. Minorsky-[C.E. Bosworth]) 
al-SAN c ANI, c Abd al-Razzak b. Hammam b. 
NafiS Abu Bakr al-Yamani al-Himyarl, Yemeni 
scholar, b. 126/744, d. in the middle of Shawwal 
211/middle of January 827. The nisba al-Himyarl in- 
dicates that he was a mawla [q. v. ] of the Banu Himyar. 
According to the biographical sources, he was of Per- 
sian origin (min al-abnP [see abna 3 ], <Idjlf, TaKkh al- 
thikat, Beirut 1984, no. 1000). His father was already 
a learned man and traditionist (Ibn Sa c d, v, 399). 
c Abd al-Razzak received his training as a scholar in 
San c a 5 , where he studied for a period of about eight 
years with Ma c mar b. Rashid (d. 153/770) (Ahmad b. 
Hanbal, al-Hlal wa-ma c rifal al-riajal, Beirut 1988, ii, 
no. 2599; Ibn Abl Hatim, al-Qiarh wa 'l-ta'-dil, 
Haydarabad 1952, iii, 38) who was himself of Basran 
origin and who had settled in San c a' after studying in 
Basra, Medina and Mecca. Moreover, c Abd al- 
Razzak participated in the lectures of a number of 
visiting scholars including the Meccans Ibn Djuraydj 
(d. 150/767), Sufyan b. c Uyayna (d. 198/813-4) and 
the Kufan Sufyan al-Thawri (d. 161/778) (Ibn Abl 
Hatim, Takdimat al-ma'-rifa li-kitab al-dj_arh wa-'l-ta c dil, 
Haydarabad 1952, 52-3; al-Dhahabl, Mizan al-iHidalfi 
nakd al-ridjal, Cairo 1907, ii, 127). It was during his 
commercial trips to Syria (al-Dhahabl, Huffaz, i, 364) 
and during the Hadjdi that c Abd al-Razzak came into 
contact and studied with other eminent scholars of the 
middle of the 2nd/8th century like al-Awza c i (d. 
157/774) and Malik b. Anas (d. 179/795) [q.vv.]. 

In the last quarter of the 2nd century A.H., c Abd 
al-Razzak became the leading scholar of the Yemen. 
His fame attracted students from all parts of the 

Islamic world, amongst them the 'Irakis Yahya b. 
Ma c In (d. 233/848) and Ahmad b. Hanbal (d. 
241/855), two famous hadith scholars of the 3rd/9th 
century. The reputation of c Abd al-Razzak was first 
of all based on his book or books. The oldest informa- 
tion about his works comes from his pupils mentioned 
above (Ibn Hanbal, Hlal, iii, no. 3882, iv, no. 3940). 
There is a difference of opinion as to the titles of his 
book or books. Ibn al-Nadlm (d. 385/995), for in- 
stance, mentions a Kitdb al-Sunanfi 'l-fikh and a Kitab 
al-Maghazi (Fihrist, Cairo 1929-30, 318), Ibn Khayr 
(d. 575/1179) claims to have known several riwdydt of 
a work called Musannaf— which is probably identical 
with the Kitab al-Sunan quoted by Ibn al-Nadim — and 
the Kitab al-Maghazi, which according to him was 
originally a part of the Musannaf. He further refers to 
a book annexed to the Azu;anna/called Djami^ which he 
describes as a work of Ma'mar b. Rashid merely 
transmitted by c Abd al-Razzak (Fahrasa, Saragossa 
1894, 127-30). Al-Dhahabl (d. 748/1347) mentions a 
Djami'al-kabir (Mizan, ii, 126), al-Safadi(d. 764/1362- 
3) a Tafsir (Nakt al-himyan, Cairo 191 1, 192), and Ibn 
Kathlr (d. 774/1372-3), finally, alongside the Musan- 
naf names a Musnad (al-Bidaya wa 'l-nihaya, ix, 265). 

c Abd al-Razzak's reputation as a traditionist is am- 
bivalent. The reasons for reservations against him 
were: he sometimes transmitted from memory and 
then made mistakes (al-Safadi, Nakt); at the end of his 
life he became blind and was then not able to per- 
sonally verify transmissions from him (al-Dhahabi, 
Mizan, ii, 127; Ibn Hadjar, Tahdhib al-tahdhtb, 
Haydarabad 1907-9, vi, 312); and he was allegedly a 
partisan of the Sh^a (Ibn Hanbal, Hlal, ii, no. 1545; 
Ibn Hibban, al-Thikdt, Haydarabad 1982, viii, 412). 
Despite these insinuations, leading traditionists such 
as Ahmad b. Hanbal, al-Bukhari. Muslim, etc., ac- 
cepted a large part of his transmission as being 
reliable, and thus c Abd al-Razzak became a key figure 
in the asdntd [see isnad] of the most important Sunn! 
hadith compilations. 

Of c Abd al-Razzak's works the following have been 
preserved: (1) A compilation of traditions from earlier 
scholars, TaW-un, Sahaba and the Prophet, called al- 
Musannaf (ed. in 11 vols. Hablb al-Rahman al- 
A<zaml, Dabhel-Beirut 1970-2; 2 1983). It contains a 
Kitab al-Maghazi and at the end a Kitab al-Djami'- which 
are both to a very large extent works of his teacher 
Ma c mar b. Rashid as transmitted by c Abd al-Razzak. 
(The ms. of a Kitab al-Salah recorded by F. Sezgin, 
GAS, i, 99, is a fragment of the Musannaf) The edition 
of the Musannaf is based on different transmissions of 
the work which have been put together (see H. Mot- 
zki, Die Anfange der islamischen Jurisprudenz, Stuttgart 
1991, 53-6). (2.) The Sahifat Hammam b. Munabbih, a 
collection of 137 ahadith of the Prophet from Abu 
Hurayra transmitted by c Abd al-Razzak through 
Ma c mar through Hammam b. Munabbih (d. 
101/719-20 of 102/720-1) (ed. Muhammad 
Hamldullah 1953, rev. ed. Luton 1979). (3.) A Tafsir 
(ms. Cairo, Dar al-Kutub, no. 242 tafsir), which is 
largely a tafsir by his teacher Ma c mar. 

The works of 'Abd al-Razzak are extremely impor- 
tant for the study of early Islamic jurisprudence, hadith 
and exegesis of the Kurgan because they contain older 
sources or materials which have otherwise been lost. 
c Abd al-Razzak had direct access to authors of the first 
extensive compilations of traditions arranged accord- 
ing to subject (musannafat [see musannaf]) like those 
by Ma'mar b. Rashid, Ibn Djuraydj, Sufyan al- 
Thawri and Sufyan b. c Uyayna. His own Musannaf is, 
to a large extent compiled from materials received 
from these four scholars, and it is very probable that 

these materials came for the most part from their 
books. In general, c Abd al-Razzak's transmission 
from these teachers of his seems to be reliable (see 
Motzki, op. cit., 56 ff.). Access to the oldest compila- 
tions of legal traditions which are not limited to pro- 
phetic hadith— as <Abd al-Razzak's Musannaf offers- 
opens new venues for researching both the origins of 
Islamic jurisprudence as well as the development of 
hadith in general. 

Bibliography: Some information on c Abd al- 
Razzak is to be found in almost all of the 
biographical sources on scholars of the 2nd and 3rd 
centuries A.H. In addition to the works mentioned 
in the article, see also Bukharf, K. al-TaMkh al- 
kablr, Haydarabad 1941-2, iii/2, 130; 'Ukayll, K. 
al-Du<afd> al-kabir, Beirut 1984, iii, 107-11. The 
most comprehensive collections of traditions about 
him are in: Dhahabi, Siyar aHdm al-nubald'- ', Beirut 
1981 ff., ix, 563-80; Ibn Manzur, Mukhtasar ta'rikh 
madlnat Dimashk li-Ibn c Asdkir, Beirut 1984 ff., xv, 
97-107; MizzI, Tahdhib al-kamdl fi asmd> al-ridjdl, 
Beirut 1992, xviii, 52-62. For the edition of the 
Musannaf mentioned in the article, two different in- 
dices are available: Fahdris, Beirut 1987 in 1 vol.; 
Films ahddilh wa-dthar, Beirut 1988 in 4 vols. 

(H. Motzki) 
al-SAN c ANI, Diva 5 al-Din Sha'ban b. Sallm b. 
c Uthman al-Hasikl al-Ruml, Yemenite poet and 
physician (1065-1149/1655-1736). 

Sha'ban's father was an Ottoman trooper from the 
town of Hasaka on the river Khabur [q.v.], and his 
mother was Yemeni by birth. Sha'ban was held in 
high esteem by his contemporaries, above all in his 
capacity as a panegyrist and man of letters but he was 
also known for his stirring sermons of exhortation. 
For some time, he lived by trade, but later he devoted 
himself to medicine and henceforth earned his 
livelihood out of medical practice (tatabbub). In his fif- 
ties, he was struck by hemiplegia (fdlidj) which in- 
creasingly isolated him as he grew older. A poor man, 
he died in his home town San c a 3 [q.v.] aged 81. 

Most of Sha'ban's ceuvre fell into oblivion after his 
death, including an anthology entitled Diwdn Badr 
Sha'-bdn. Except for some fragments of poetry as pre- 
served by his biographers, only two pieces from his 
pen have come down to us, sc. a dispute about rank 
between a freeborn woman and a female slave (see 
Bibi), and an extensive didactic poem on dietetics and 
hygienics designed for domestic use (see Bibl.). 

Bibliography: O. Kahl, al-Mufahara baina l-hurra 
wal-ama. Ein Beitrag zur jemenitischen Rangstreitliteratur 
des 17.118. Jahrhunderts, in WO, xvii (1986), 110-49; 
A. Schopen and O. Kahl, Die NatdHg al-fikar des 
Sa < bdn ibn Sallm as-San^dni. Eine jemenitische Gesund- 
heitsfibel aus demfruhen 18. Jahrhundert. Text, Uberset- 
zung und Kommentar, Wiesbaden 1993, 3-8 (and the 
bio-bibliographical sources quoted there). 

(O. Kahl) 
al-SANAWBARI, Abu Bakr (or Abu '1-Kasim or 
Abu '1-Fadl) Muhammad b. Ahmad (or Ahmad b. 
Muhammad) b. al-Husayn b. Marrar al-Dabbl (the 
nisba al-Slni, given by Mez, Renaissance, 250, Eng. tr. 
261, is a scribal error) al-Halabl al-Antakl, poet and 
librarian at the court of Sayf al-Dawla [q.v.] in 
Mawsil and Aleppo. 

Born in Antioch before ca. 275/888 (if Ihsan 
c Abbas, Diwdn al-Sanawbari, Beirut 1970, poem 231, 
v. 67 is to be taken literally), he died in Aleppo in the 
year 334/945. Since his chronology is consonant with 
neither the Aghdni nor the Yatima, very little is known 
about his life. He was a close friend of Kushadjim 
[q.v.], the poet and master-cook to Sayf al-Dawla, and 

eventually became his son-in-law, exchanging several 
verse epistles with him, as was the practice of the time 
(see J.E. Montgomery, Abu Firas's poetic correspondence 
with Abu Zuhayr, in The Occasional Papers of the School of 
Abbasid Studies, St. Andrews, ii [1988], 1-45). The 
tradition that al-Sanawbari was himself a keen 
gardener may be aetiological, based on his fame and 

s Mez, 

plains, the lakab a, 
either he or his father traded in pine nuts (sanawbar). 
Al-Sanawbari, however, according to Ibn c Asakir, 
Tahdhib Ta>rfkh Dimashk, Damascus 1329, i, 456, 
claimed that the name had been given to his grand- 
father by the caliph al-Ma'mun [q.v.], on account of 
the keenness of his reasoning in disputations (?). His 
poems in praise of the ahl al-bayt [q.v.] point to his 
Shr 5 ! affiliations, but as this is not attested by any later 
authorities it may simply be that these poems were 
composed to harmonise with, or on behalf of, the Sh?! 
Hamdanids [q.v.]. Among his rawis were Abu '1- 
Hasan al-Adib and Abu '1-Hasan Muhammad b. 
Ahmad b. Djumay c al-Ghassani. 

Al-Sanawbari's diwdn was edited (in alphabetical 
and not generic order) by Abu Bakr al-Suli [q. v. ] . Ac- 
cording to the Fihrist, 168, it amounted to 200 folios. 
The section containing the rhyme-letters ra 3 to kdf is 
still extant in a ms. in Calcutta (see c Abbas, op. cit. , 
6), although this is not exhaustive as c Abbas, Takmilat 
Diwdn al-Sanawbari, poems 83-98, and L. al-Sakkal 
and D. al-Khatib, Tatimmat Diwdn al-Sanawbari, Alep- 
po 1971, poems 27-37, indicate. The diwdn contains 
the full panoply of poetic types encountered in the col- 
lections of other contemporary poets (panegyrics and 
vituperations [50 pieces], love poems [especially 
mudhakkardt, poems composed about boys, 80 pieces], 
veneric and cynegetic pieces, threnodies and consola- 
tions [40 pieces], occasional snippets composed to 
order, etc.) but is justly famous for its 40 nature 
poems, both epigrams and fcwwijs [l ■"■]'■ 'abi^iyydt 
(vernal poems), zahriyydt (floral poems), rawdiyydt 
(meadow poems) and even thaldpyyat (snow poems). 
Whilst al-Sanawbari is heavily indebted to his poetic 
forbears, especially Ibn al-Mu c tazz [q.v.], he is 
acknowledged as the creator of the fully independent 
nature kasida, doing as much for horticultural poetry 
as Abu Nuwas [q.v.] did for viticultural poetry. His 
achievement is that he not only liberated nature as a 
theme of the Arabic poetic repertoire, establishing it 
as a genre in its own right, but that he also enlarged 
its scope to encompass other genres, such as the 
khamriyya [q.v.]. The key-note of this poetry is the 
striving after wonderment and effect, being very 
much the product of the badi* [q.v.] style and being 
heavily dependent upon "phantastic" (takhyTli) im- 
agery, as later defined by c Abd al-Kahir al-Djurdjani 
(d. 471/1087 [q.v. in Suppl.]), imagery which involves 
the inversion of the conventional leading to an an- 
thropomorphisation of nature. These effects are 
directed at the auditor/reader's appreciation of the 
poet's display of "linguistic ingenuity ... a style in 
which the probable is made improbable, the familiar 
enigmatic, the ordinary miraculous" (S. Sperl, Man- 
nerism in Arabic poetry, Cambridge 1989, 156). 

In modern scholarship, al-Sanawbari's nature 
poetry has been studied from the exclusive viewpoint 
of the imagination, as literary (mannerist) artifice 
devoid of any connection with reality. This is to 
overlook the occasional or social nature of much of 
this poetry; it is not the poetic exercise of the imagina- 
tion, but was presumably composed for the garden 
banquet-cum-m<2^/« "when the caliph or a grandee 
would invite the habitues of his salon to a Sans-souci 


in the gardens at the edge of town" (G.E. von 
Grunebaum, Aspects of Arabic urban literature, in Islamic 
Studies, viii [1969], 293). Hence the repeated, witty 
shifts in register, from the "phantastic" to the on- 
tological, i.e. from the imaginary to the real. His in- 
fluence on Andalusian poetry was such that Ibn 
Khafadja [q.v.] was crowned the Sanawbarl of the 

A commentary on Dhu '1-Rumma's BcPiyya was 
also penned by al-Sanawbarl. 

Bibliography: Given in the article. For a survey 
of 20th-century scholarship and an extensive discus- 
sion of al-Sanawbarl's nature poems, see G. 
Schoeler, Arabische Naturdichtung. Die Zahnyat, 
Rabidly at und Raudiyat von ihren Anfangen bis as- 
Sanaubari. Eine gattungs-, motiv- und stilgeschichtliche 
Untersuchung, Beirut 1974, 273-343; and Sezgin, 
GAS, ii, 501-2. Other works containing discussions 
of al-Sanawbari are A. Hamori, On the art of medieval 
Arabic literature, Princeton 1974, 78-87; M.M. 
Badawi, From primary to secondary Qasidas: thoughts on 
the development of Classical Arabic poetry, in JAL, xi 
(1980), 29-31; E. Wagner, Grundzuge der klassischen 
arabischen Dichtung, Darmstadt 1988, ii, 145-50; J. 
Stetkevych, The zephyrs of Najd: the poetics of nostalgia 
in the Classical Arabic Nasib, Chicago 1993, 183-7. 
See also C.E. Bosworth, Sanawbari's elegy on the 
pilgrims slain in the Carmathian attack on Mecca 
(317/930): a literary-historical study, in Arabica, xix 
(1972),_222-39. (J.E. Montgomery) 

SANDABIL, a town said to be the capital of the 
king of China in the account of the Arab traveller 
and litterateur Abu Dulaf Mis'ar b. Muhalhil [q.v.\ 
purporting to describe his participation in an embassy 
of the Chinese king Kalin b. al-Shakhlr returning 
from the court of the Samanid amir Nasr b. Ahmad 
(301-31/914-43 [q.v.]) at Bukhara. 

Abu Dulaf describes it as an immense city, one 
day's journey across, with walls 90 cubits high and an 
idol temple bigger than the sacred mosque at 
Jerusalem (First Risala, Fr. tr. G. Ferrand, in Relations 
de voyages . . . relatifs a I 'Extreme Orient du VIII' au XVIII' 
siecles, Paris 1913-14, 219-20, 221; Ger. tr. A. von 
Rohr-Sauer, Des Abu Dulaf Bericht iiber seine Reise nach 
Turkestan, China und Indien, Bonn 1939, 17, 27-30, text 
Originally known in Yakut, Bulddn, ed._ Beirut, iii, 
440, 444-5, cf. 275, and in al-KazwTm, Athdr al-bildd, 
ed. Wiistenfeld, ii, 30 ff.). It is not, however, men- 
tioned in other early Muslim sources on China (the 
Hudud al- c alam, Gardlzl, MarwazI), and its existence 
as such is dubious, given also further doubts about the 
historicity of this Sino-Samanid mission (see C.E. 
Bosworth, An alleged embassy from the Emperor of China to 
the Amir Nasr b. Ahmad: a contribution to Samanid military 
history, in Minovi and Afshar (eds.), Yddndme-ye irani- 
yeMinorsky, Tehran 1969, 8-9). Marquart, Streifzuge, 
84-9, cf. von Rohr-Sauer, op. cit., 58-60, sought to 
identify Sandabil with Kan-chu, in Kan-su province 
[q.v.], which would be geographically feasible. Minor- 
sky pointed out, Hudud al- c dlam, comm. 232, that the 
name of the "large town governed from China" in 
the anonymous geography, tr. 85, Khal.b.k. 
resembles in the Arabic script the S.n.dab.l of Abu 
Dulaf, especially in its last three letters, and noted that 
the place mentioned in the Hudud al-'alam 's chapter on 
China immediately before Khal.b.k is Khamcu = 
Kan-chu. It seems impossible to take the question any 
further than this. 

Bibliography: Given in the article. 

(C.E. Bosworth) 
SANDAL (a. , P. candal from Skr. candana) is the 
sandal wood, coming from several unrelated trees 

which are mainly of Indian and Southeast Asian 

Both white and yellow sandalwood were, in fact, 
only different kinds of Santalum album L., Santalaceae. It 
supplies the bright, white sap-wood and the reddish 
heartwood. Because of its peculiar scent, probably ex- 
perienced as very pleasant, it was appreciated from 
time immemorial and used, among other purposes, 
for perfumeries, and its ethereal oils against inflam- 
mations of the urinary passages. The red sandalwood, 
on the other hand, is the heartwood of Pterocarpus san- 
talinus L. , Leguminosae. It is totally scentless and of lit- 
tle value, but was popular for its beauty. It is not 
known how the name sandal was transferred from the 
white-yellow to the red wood. The Arabic authors 
know the same threefold distinction. Sandalwood was 
unknown to the Greeks. The most comprehensive ac- 
count about sandal is found in al-Nuwayrl, Nihaya, xii, 
39-42, tr. Wiedemann, in Aufsdtze zur arabischen 
Wissenschaftsgeschichk, ii, 252-4, 263. The yellow, fat 
(al-dasim), heavy wood, which looks as if it were 
painted over with saffron and is therefore also called 
al-zafardni, is accounted the best sandalwood. It has a 
strong fragrance and is designated as al-makasiri (the 
meaning of this nisba is not clear, cf. Dozy, Suppl. , ii, 
358-9). Of the white sandalwood, which is also 
fragrant, there exist various varieties, which incline 
partly to yellow, brown and red. 

The use of various sandalwoods in medicine, above 
all of their ethereal oils, is described extensively from 
several sources by Ibn al-Baytar (Djami c , iii, 89, 10- 
31 = Leclerc, no. 1418). When added to electuaries 
(ma c djundt), they are inter alia effective against fever 
and heating of the bile. If inhaled, its powder is effec- 
tive against pleurisy (birsdm) and congestion (? lahib). 
If the electuary is applied together with rosewater as 
a poultice, it is effective against erysipelas (humra), 
boils of feverish gout (al-nikris al-hdrr) and infections of 
the eyelids (shatar). 

In the Maghrib, as repeatedly stated by c Abd Allah 
b. Salih, Ibn al-Baytar's teacher (see Dietrich, 
Dioscurides triumphans, iii, 31, 38, 93, and Dozy, Suppl. 
i, 846a), sandal indicates thyme (nammdm) and 
the wild and cultivated mint (Thymus serpyllum L., 

Bibliography: Studies of the sources and further 
literature in A. Dietrich, Disocurides triumphans. Ein 
anonymer arabischer Kommentar, in Ende 12. Jahrh. zur 
Materia medica, i, 41, n. 2, Abh. A.W. Gott., Phil- 
Hist. Kl, N.F., no. 173, Gottingen 1988; idem, Die 
Ergdnzung Ibn Gulgul's zur Materia medica des 
Dioskurides, no. 38, in Abh. A.W. GStt., Phil.-Hist. 
Kl, N.F., no. 202, Gottingen 1993. 

(A. Dietrich) 
SANDJ. Sindj, pi. sunudi, the generic term for any 
kind of cymbal. Both al-Djawhan and al-Djawallki 
say that the word is an Arabicised one. Lane thinks 
that it is derived from the Persian sandj_ or sindj_ and 
Ibn Khurradadhbih (d. near the opening of the 10th 
century) avers that the Persians invented it (al- 
Mas'OdT, Murudx, viii, 90 = § 3214). However, the in- 
strument was well known to the ancient Semites. We 
read of the sandj_ in early Arabic literature. Al-Kutami 
refers to the sandj_ al-djinn and Ibn Muhriz [q. v. ] was 
called the sannadf al-'-Arab. The feminine form of the 
latter, said to express an intensive, is also to be found 
in the cognomen of al-A c sha Maymun known as the 
sannddjat al-'-Arab and in a certain Mustarad al- 
sannddja. Yet it is difficult to say whether the actual in- 
strument or mere symbolism is aimed at in these in- 
stances. Further confusion is added by the fact that 
the word sandf (< Pers. cang) was also given by some 

Arabic writers to the harp, although the more general 
name for the latter was djank [see mi'zaf]. 

The term sandj or sindj is generally used for the cym- 
bal in the East, although zindj has been more common 
in the West since the Middle Ages. The instrument is 
played in pairs and is used to regulate the measure or 
rhythm in both music and dancing. That it had a 
definite place as a rhythmic intrument in days of old 
is stated by Ibn Zayla (d. 440/1048) in his Kitab al-Kafi 
(ed. Z. Yusuf, Cairo 1964). It is to be found in several 
shapes and sizes. The finger cymbals used today are 
generally about 4 or 5 cm in diameter and they are 
usually attached to the thumb and middle finger. 
They are depicted by Niebuhr (i, tab. xxvi), Villoteau 
(pi. cc. 26), Lane {Modern Egyptians, ch. xviii), Christ- 
ianowitsch (no. 36), Lavignac (2794, 2936), and 
Sachsse (tab. 8, no. 36). Specimens may be found in 
museums, notably Brussels (no. 293) and New York 
(no. 383). Other names for the cymbal, according to 
Villoteau (980), are zil (< Turk, zill), kds (probably 
of cup-shape form originally), and sadidja or sadidja. 
although probably this ought to be written sadidja. In 
Syria we have the term fukaysha, and in Morocco 
nuwayksa (dimin. of nakus) in common use, the former 
being a metathesis of shukayfa (see below). The term 
saldsil (sing, salsal) was also applied to all high- 
sounding clashed metal instruments, of this type. Like 
zil or zill, it is of onomatopoeic origin, the verbal root 
being salla ("to sound"). There are cognates in all the 
Semitic languages. Saadia (d. 941) equates the Arabic 
root with the Hebrew sdlal, and we have the Arabic 
musalsaldt standing for the Hebrew self Urn (cymbals) 
of Psalm cl. 5, in the Glossarium latino-arabicum (11th 
century). Small cymbals attached to a frame were also 
in use. This instrument was known as the djaghdna or 
saghdna (see below). It resembled a pair of metal tongs 
with two or three arms branching from the open ends, 
a small cymbal being attached to each arm. Nowadays 
it is called a zilli mdsha ("jingling tongs"). We see it 
depicted in Sasanid art, and it is mentioned by Ibn 
Khallikan (tr. de Slane, iii, 491) and in the Anwdr-i 
Suhayli. There are two Turkish specimens at New 
York (nos. 353, 1377). 

The hand cymbals are to be found in both the 
plate and bowl shape. This belongs to martial and 
processional music. Clement of Alexandria 
(Paedagogus) said that the Arabs used cymbals (xu|x- 
PaXoc) in war, and this seems to be hinted at in the later 
Arabic reference to the sannddjat al-djaysh, although 
Arabic lexicographers think differently. Al-Djawhari 
describes a cup-shape instrument called the sahn. It 
was a small bronze cup (tusayt) which was struck 
against another of its kind. This cup or bowl-shape 
cymbal was favoured in martial music, and it is 
delineated in several pictures of a military band which 
are found in the treatise on automata by Badl c al- 
Zaman al-Djazarl (flor. later 6th/ 12th century [q.v. in 
Suppl.]) which have been reproduced (The legacy of 
Islam, lsted., fig. 91; Schulz, Die pers. -islam. Miniatur- 
malerei, tab. ii; al-Djazari, tr. D.R. Hill, The book of 
knowledge of ingenious mechanical devices, Dordrecht 
1974). At this period, however, the cymbal was called 
the kds, kasa or ka's, and Nasir-i Khusraw (Safar-nama, 
ed. Schefer, 43, 46,47) mentions it among the martial 
instruments of the Farimids. In the Alflayla wa-layla (i, 
66,323; ii, 656; iii, 150, 271, 274, 298), these bowl- 
shaped ku 5 uj or kdsdt are frequently mentioned in com- 
pany with tubul (drums) in the warlike scenes. In 
modern times, the hand cymbal is plate-shaped and 
known as the sana^j zil and kds (Villoteau, loc. cit. ; 
Russell, Aleppo, i, 151). Villoteau gives the diameter 
of the Egyptian instrument as 24.4 cm. For a Palestin- 

ian example, see Sachsse (66, tab. 8). For numbers 
used in military bands, see tabl-khana. For quite a 
century and a half, Turkey has been famed for the 
manufacture of cymbals and in the earlier part of the 
20th century several thousands were exported from 
Istanbul every year. There are two other mediaeval 
names for the cymbal which are worth recording, viz. 
saffdkatdn and musdfik. The former occurs in the Kitab 
al-Aghdni(v, 75), and Ibn Hadjar al-Haytaml (Berlin 
ms. 5517, fol. 19b) likens it to the sandj (cymbal). 
Musdfik and musdfika equate with cymbalum in the 
Glossarium latino-arabicum and the Vocabulista in Arabico 
(12th- 13th century). 

Clappers. In Arabic, handclapping is called safk, 
safk, tasfik, tasfik and tasfih, all of these terms being 
derived from verbal roots meaning "to clap the 
hands", and are of the same kin as the Hebrew sdpak 
(Ezekiel, xxi. 17). A plate of wood or metal was called 
a safiha, and from the same root we get musaffahat, a 
word which appears to denote "clappers". Labld 
[q.v.], the Arabic poet, places musaffahat in the hands 
of wailing women (anwdh). Another word for clappers 
occurs in the Vocabulista Aravigo (1501) where we have 
maciquif (chapas para taner) and mabiquif (tarrenas chapas 
para taner) registered. Doubtless the b in the latter is a 
slip for c. Dozy was of the opinion that both these 
words were metatheses of musdfik, but it is more likely 
that the word intended is mashdkif (sing, mishkifa), the 
Aramaic root corresponding to this being shekaf ("to 
clap the hands"). See also shakf and shukuf (testa) in the 
Glossarium latino-arabicum and the Vocabulista in Arabico. 
In modern times the shukayfdt were small small cym- 
bals (or castanets) used by dancers. For a design of 
these clappers, see the Kitab al-Burhdn in the Bodleian 
Library (Or. 133, fol. lib). In Persia and Turkey 
they are known as the corpora (lit. "four pieces") or 
calpdra. See Farmer, Turkish musical instruments, in 
JRAS (1936). Castanets are mentioned by Ibn 
Khaldun (ed. Quatremere, ii, 354, tr. Rosenthal, ii, 
397) and Villoteau (981) says that they were called 
akligh in Egypt. Outside of Spain, where they may 
have been known as the kdsatan (hence perhaps 
Castanet), they have not been favoured. 

The Percussion slab known as the nakus is dealt 
with separately; see nakus. 

Percussion staff. This was the kadib, an instru- 
ment found in the hands of several of the early musi- 
cians of Islam. Its identity has long been a puzzle to 
both musicographers and orientalists. It was a staff 
which was used for rhythmic purposes either by strik- 
ing it upon the ground or upon something else. Ibn 
Hadjar al-Haytaml (d. 972/1565, fol. 19b) has a sec- 
tion entitled "Concerning beating (darb) with the kadib 
upon cushions (wasd^id)". It recalls an incident in the 
"Story of the Mock Caliph" in the Alflayla wa-layla 
where a cushion (mudawwara) is struck as a signal for 
servants to appear. Burton will not allow that a 
cushion is meant, and substitutes "a circular plate of 
wood or metal, a gong". We get a slight idea of the 
sound of the kadib from the fact that Muhammad is 
said to have been averse to the tick-tack (taktaka) of the 
kadib, and the same is said of the imam al-Shafi c I (al- 
Shalahl, fol. 79). It is given a place in music by the 
Ikhwan al-Safa' (i, 91) and Ibn Zayla, although later 
it fell into desuetude and was only to be found in the 
hands of the amateur and the folk. Indeed, the word 
muktadab came to mean "untrained" or "extem- 

Bells. Ordinarily, the cup, bowl, or cone shape 
bell is known in Arabic as the djaras, whilst the sphere- 
shape bell is called the djuldjul. On the other hand, 
djaras also stands for a large bell (campand) and djuldjul 


for a small bell (tintinnabulum), the probable reason 
being that the first-mentioned form was generally 
found in the large instrument whilst the second- 
mentioned form was generally found in the small in- 
strument. Bells were used on the necks of animals in 
pre-Islamic days, and there is a tradition that 
Muhammad was averse to the sound of the caravan 
bells so that the fiction arose that "angels will not 
associate with a company where there is a paras' '(e.g. 
Muslim, Libds, trad. 103). A collection of these bells, 
either on a board or on a chain or rope, is known as 
a fabla. The term was probably borrowed from the 
Hebrew fabla, which, in turn, had its origin in the 
Greek totpXot, because these bells were generally at- 
tached to a tablet of wood. There is a specimen of a 
labia at New York (no. 2659), the largest bell being 
10 x 5.8 cm. Bells were also used to increase the din 
of battle so as to affright the enemy, as we are told by 
Ibn Zayla, and in the story of Gharlb and his brother 
c AdjIb in the Alflayla wa-layla (iii, 294) we read of the 
camels and mules in battle being furnished with large 
bells (adjras), small bells (djaladjil), as well as jingles 
(kalakil). According to Cervantes, the Moors of Spain 
did not tolerate their use as martial instruments. 

The small bell (djuldjul), sometimes called a pellet 
bell, was spherical. Like salsal, dabdab, etc., the word 
is of onomatopoeic origin. Al- Kh alll (d. probably in 
175/791) likened the sound of the small cymbals 
(sunudj) hanging in the rim of the tambourine (duff) to 
that of the small bells (djaladjil; see Kh"arazmi, 
Mafatih al-^ulum, 236). Indeed, these small bells were 
sometimes attached to tambourines [see duff]. Al- 
Muzarrid (6th century A.D.) speaks of small (tam- 
bourine) bells (djaladjil) replying to the wind in- 
struments (mazamir; see the Mufaddaliyyat, i, 165). 
These djaladjil were also attached to the necks of 
smaller animals in the form of a tab la, and in Mamluk 
times they were fastened to the hats of criminals (al- 
MakrizI, Suluk, i/2, 106). They also formed part of the 
impedimenta of itinerant minstrels, who likewise wore 
them on their hats (J.S. Buckingham, Travels, i, 100), 
as did the fools in Talmudic Jewry (Jastrow, Diet. 
Targ. , 518). In Persia, the large bell is called a zang or 
dara and the small bell a zangula or zangulica. In Turkey 
they are the c'ang and cingrak respectively. 

An elaborate type of chimes was known to the 
Arabs, who borrowed the idea from the Greeks. It is 
described in a treatise by one Muristus [q.v.], who, in 
turn, was indebted to an Egyptian named Sa c atus or 
Safus, whose writings were known in Arabic as early 
as the 4th/10th century at least (Fihrist, 270). This in- 
strument was called the djuldjul al-sayyah (clamorous 
bell) or the djuldjul al-siyah (octavo bell). See Maeh. , ix, 

Another jingling instrument was the djaghana or 
saghana (< Pers. caghand). It took several forms. One 
was a sceptre of wood surmounted by hoops of wire 
from which were suspended about a hundred small 
bells. For a design, see Niebuhr(tab. xxviii). Another 
kind was surmounted by a metal cone pavilion, hence 
the European name of Chapeau Chinois which was 
given to it. From this, and from three or four horizon- 
tal arms, small bells and cymbals were hung. It was 
borrowed by European military bands in the 18th cen- 
tury from the Turks, and in Britain was known as the 
"Jingling Johnnie". See Farmer, Rise and development 
of military music, fig. 9. For the Turkish instrument, 
see Wittman, Travels in Turkey (1803). Oriental Chris- 
tians use a different type known as the mirwaha (lit. 
"fan"). It is described and delineated by Bonanni 
(127, pi. lxxxix), La Borde (i, 282), and Vil'loteau 
(1008-10). A fourth type is the dabbus used by the dar- 

wish fraternities. It is a wooden sceptre, to the head of 
which is attached a number of chains (saldsil) with 
jingling pieces of metal fixed loosely in the links. 
There is a specimen 69 cm long at New York. 

Rattle. This is generally known as the shakhshikha. 
In Persia and Turkey, there is the kashik, which is two 
wooden spoons attached to each other, in the hollow 
of which are a number of small bells. It is more 
generally struck with a stick. See Advielle (15) and 
Lavignac (3076). 

Harmonica and glockenspiel. The Ikhwan al- 
Safa 3 (i, 90) deal with vessels (awdni), pots (larajahdrat), 
and jars (djirar) as idiophones. In Arabic, the general 
name for the harmonica was tusit and Ibn Khaldun 
speaks (loc. cit. ) of these tusut being played with sticks 
(kudban). The Persian Ibn Ghaybl describes saz-i kasat 
(lit. musical bowls), which were made of earthenware 
and the notes of which were determined by the 
amount of water with which each bowl was filled 
(Djami" al-alhan, Bodleian Library, ms. Marsh 282, 
fols. 78, 81b). An Arabic author of the 9th/15th cen- 
tury refers to the harmonica as the kizan (cups) and 
khawabP (jars) and mentions the water content (B.L., 
Or. 2361, fol. 173). Ibn Hadjar al-Haytami describes 
(fol. 19b) the beating with reeds (aklam) upon earthen- 
ware (sini). The glockenspiel is mentioned only by Ibn 
Ghaybl (fol. 81b), and he registers the instrument 
under saz-i alwdh-i fuldd ("instrument of slabs of 
steel"). It comprised 35 slabs, each giving a particular 

Bibliography: See that to tabl, and add 
Sachsse, in ZDPV (1927); La Borde, Essai sur la 
musique ancienne et moderne, 1870. 

(H.G. Farmer) 
SANDIA, the name of a small, right-bank af- 
fluent (Grk. Singas, Modern Tkish. Keysun Cayi, a 
tributary of the Gok Su) of the upper Euphrates and 
of a small town on it, both coming in mediaeval 
Islamic times within the northern part of Diyar 
Mudar [q.v.]. The Sandja river runs into the 
Euphrates between Sumaysat and Kal c at al-Rum 
[q. vv.]. It was famed for its bridge, said by the Arabic 
geographers to have been composed of a single arch of 
200 paces' length constructed from dressed stone, and 
to have been one of the wonders of the world (cf. 
Yakut, Buldin, iii, 264-5). It was here and at nearby 
Baddaya that the Artukid Nadjm al-DIn II Ghazi 
crossed in 513/1119 on his campaign at Tell Bashir 
against the Franks from the County of Antioch (see 
Cahen, La Syrie du Nord, 283 ff., and mardj dabik). 
Bibliography: Le Strange, Palestine under the 
Moslems, 531; idem, Lands, 123-4; CI. Cahen, La 
Syrie du Nord a I'epoque des Croisades, Paris 1940, 127, 
295-6; J. Tischler, Kleinasiatische Hydronymie , Wies- 
baden 1977, 136. (C.E. Bosworth) 
SANDIAK (t.), a Turkish term with various 
significations. (1) flag, standard, banner (Arabic 
tiwa y ), especially of a large size (more important than 
the bayrak, Ar. raya or < atam) and suitable for fixing in 
the ground or hoisting permanently on a monument 
or a ship; (2) (nautical term) ensign; pennant (ikindji 
sandjak), starboard; (3) formerly a military fief or khass 
[q.v.] of a certain extent in the Ottoman empire; (4) 
a Turkish administrative and territorial division; (5) 
(in the expression sandjak tikeni or dikeni, from the 
Turkish translation of Burhan-i kali*, 88, 25) a 
synonym of sindjan tikeni (on this plant, see Barbier de 
Meynard, ii, 101, who gives it as a Persian word). 

As al-Kalkashandl pointed out in the 9th/15th cen- 
tury (Subh al-a'-shd, v, 458), sandj-ak comes from the 
verb sandj-mak (not sandji-mak, as in the author already 
quoted) which means "to sting, prick, plant, stick a 

weapon or pointed object in the body of an enemy or 
in the ground (cf. Sam! Bey, Kdmus-i TurkT). The form 
sancak found in Ca gh atav (Boudagov) and even in an 
old Serbian loanword (Miklosich, Die liirkischen 
Elemente in den sudost-europaischen Sprachen, Vienna 
1884, ii, 50) corresponds to the verb sane- of the Or- 
khon inscriptions (see Thomsen, 42; Radloff, 132). 
Cf. also F.W.K. MQller, Uigurica, ii, 78, 30 and 86, 
48. In Kirghiz the form used is shansh- (Radloff, 
Worlerbuch, iv, 949), and in Uriankhay shanish- and 
canish- (Katanov, Opit izledovaniya, 429, 779, with the 
meaning "to prick, stab, erect, fix"). Mahmud al- 
Kashghari (5th/llth century), Diwdn Lughdt al-Turk, 
ed. Kilisli Rifat Bey, ii, 171, 180, 182, iii, 310, also 
gives (iii, 108) sandjghan equivalent to sandjan (sindjan) 
already quoted, which is a Turkish participle used as 
the name of a prickly plant. 

The word sandjak belongs to a family of derivatives 
which all contain the idea of "point" and mean (the 
word itself sometimes): harpoon, fork, piercing pain, 
colic. Such are sancigh, sandjikh, sandjlk, canckl 
(Tobolsk), shanhhki (Kirghiz), sandjighi, sandjl (whence 
sandji-mak in c Othmanli). We may add on the authori- 
ty of Abu 'l-Fida' and the Turkish-Arabic glossary 
published by Houtsma, Leiden 1894, 80, and 29 of 
the Arabic text, the proper name Sandjar [?.».], gloss- 
ed yafan, in preference to the usually accepted etymo- 
logy from Sindjar, the name of his place of birth (see 
Recueil des historiens des Croisades, i, 1872, and index 
under Sindjar). 

Sandjak has passed into a certain number of other 
languages; more recently into the Balkan languages 
(cf. the work by Miklosich quoted above and Sai- 
neanu, Influenta orienlala), and earlier into Arabic (cf. 
Dozy, Supplement, and W. Marcais, Le dialecte arabe de 
Tlemcen, Paris 1902, 270, 90, 92), and into Persian 
where, according to the Burhdn-i kdli c , it means or 
meant a "flag, a large metal pin intended to keep on 
the head a kind of hood worn by women"; "a kind of 
girdle". In Modern Persian sandjak (sic) simply means 
"pin" (in opposition to "needle") (cf. Nicolas, Dic- 
tionnnaire jrancais-persan, under the word "pin"). 
Freytag took sandjak for a Persian word, and the Turks 
still keep the orthography which it has in Persian 
(s-n-dj-d-k), while they write the verb sandj- with a sad. 
We may note that in Persian dirafsh "flag" also means 
"point" (see Vullers), whence the Ottoman Turkish 
word direwush (see Hind-oghlu s.v. "pointe" and 
"poincon"). The Burhdn-i kdff- gives a variant of san- 
djak in the form sandjuk. If it is not a corruption due 
to the Persian, we have here another example of a 
Turkish word preserved through its use in Persian. 
The word sandj-uk is very well explained with the help 
of the Turkish suffix -uk (-ik) which makes a passive 
participle from transitive verbs. Sandjuk then would 
mean "sharpened, fixed". The suffix -ak, with its 
tendency to designate place-names (which very well 
fits a flag "fixed" or "able to be fixed") seems to have 
been more in use very early. 

The etymological details which are given above 
without excluding the explanation of sandjak by "lance 
with a pennon" (it is that of al-Kalkashandi, who uses 
the word rumh) make very probable the explanation as 
"flag with a staff sharpened at the foot". In- 
dependently of this peculiarity it is difficult to say 
what was the exact form of the primitive Turkish san- 
djak; did they have a horse's tail (or the tail of a yak, 
of which von Hammer speaks in his definition, Hist, 
de I'empire ottoman, xvii, 257) or were they always flags? 
Were they like the calish or shallsh mentioned by Ibn 
Khaldun (for the references see Dozy, Supplement, 
under the word djdlish; it has become hdlTsh by an error 

in Djewdet Pasha and Ahmed Rasim, quoted below 
in the Bibliography)? The meaning of these terms may 
be more indefinite than we think, and varied a great 
deal with time and place. The word tugh [q.v.], which 
could be taken in the meaning "horse's tail", meant, 
according to al-Kashghari, not only a "flag of silk or 
orange brocade" but also "drum", another symbol of 
sovereignty (i, 169, iii, 92). Ibn Khaldun confuses the 
flag with the ' ' parasol " of the prince or djitr, better c'atr 
(Persian) pronounced callr (al-Kashghari, i, 340), 
then cadir "tent", by the Turks who have preferred 
these words to their old covac "silk parasol of the 
Turkish Khaghans" (ibid, ii, 149, 17, iii, 45, 15; cf. 
the c Othmanli coghash "a place in the sun", and a 
passage in Rabghuzi in Radloff, Wbrterbuch, iv, 59, 

Whatever its primitive form, the sandjak appears 
among the Saldjuks as an insignum of royalty. In the 
Turkish text of Ibn BIbl (ed. Houtsma, in Recueil, iii) 
the word sandjak is always found in connection with 
the title Sultdn (Sultanin sandjagju). This standard is 
mentioned (135-6, 144, 169, 170, 289, 357) a propos 
of different sieges of strong places, on the walls of 
which it was placed after capitulation. Sometimes 
(135-6) it is the besieged themselves who, ready to sur- 
render and no doubt seeing in this banner a guarantee 
of protection against pillaging, asked for a sandjak to 
be sent. It is not, however, necessary that the sultan 
himself should be present and the historian (357) 
shows us the beylerbeyi setting out on an expedition 
with the standard of the sovereign. 

For a long time the neighbouring princes and 
vassals of the Saldjuks respected their privilege, but 
the Atabeg of Mawsil, Sayf al-DIn al-GhazT, son of 
c Imad al-DIn Zangi (d. November 1 149), was the first 
of the ashab al-atraf to have a sandjak carried unfurled 
over his head (Ibn al-Athir, Hist, des Atabeks de 
Mossoul, in RHC, Hist, or., ii/2, 167). 

The Ayyubids followed the example of their 
predecessors. In 594/1198 the sultan of Egypt, al- 
Malik al- c AzIz, conferred on his nephew al-Malik al- 
Mu c azzam c Isa, when he became prince of Damascus, 
"the sandjak and the liwd* to display throughout the 
world" (Kitdb al-Rawdatayn, in RHC, v, 117). In 
648/1250 Aybak the Turkoman, married to an 
Ayyiibid princess and proclaimed sultan of Egypt, 
took part in a procession in which the royal banners 
were unfurled for him (al-sanddjik al-sultdniyya; cf. Abu 
'1-Fida>, Annates, ed. Reiske, iv, 516 of the Arabic text 
and 515 of the Latin tr.). Among the Mamluks, a 
distinction was made between the sandjakddr "royal 
standard-bearer" and the ordinary c alamddr ( M. 
Gaudefroy-Demombynes, La Syrie a I'epoque des 
Mamelouks, Paris 1923, p. xcvii); afterwards, in 
Turkish Algeria this distinction disappeared; see J. 
Deny, in Melanges Rene Basset, Paris 1923-5, ii, 35. 

According to one later Ottoman tradition, not to be 
taken literally but nevertheless attesting to the 
significance of the sandjak/banner as a political sym- 
bol, at the end of the Saldjuk empire in Asia Minor 
the sandjak became one of the insignia of investiture of 
new sovereigns, notably of the first c Othmanli sultan. 
In 679/1280, after the capture of Karadja Hisar by 
c Othman, Sultan 'Ala' al-Din II to celebrate this con- 
quest sent him by the hands of Ak Timur, c Othman's 
nephew, a sandjak "with its accessories" (sandjak 
yaraght), as c Ashik Pasha-zade tells us (ed. Constan- 
tinople 1332, 8-9); Neshri prefers another version (see 
Noldeke, in ZDMG, xiii [1859], 207-9). c Ashik Pasha- 
zade mentions in this connection that c Othman thus 
became sandjak beyi, and we know that it was from this 
time that the khutba was read in his name (for the first 


time at Karadja Hisar by Dursun Faklh). According 
to the same authority, the sandjaks were made of cloth 
produced in Philadelphia or Ala Shehir (56). 

When they became independent in their turn, the 
Ottoman rulers appointed sandjak beyis in larger and 
larger numbers and the sandjak, somewhat diminished 
in splendour, became identified with the territory over 
which it waved, not as a symbol of independence but 
of political authority deputised by the ruler; it appears 
henceforth as the term for a political/administrative 
division. The original patrimony around Bursa re- 
mained the Ottoman bey's [q.v.] domain; areas added 
by conquest, such as KarasI [go-] and Izmid [q.v. in 
Suppl.] in Anatolia or on the frontier zone in Thrace 
were entrusted, as newly created sandjaks, to other 
members of the House of c Othman or to the com- 
manders leading the conquest. In time, each of the 
smaller Anatolian principalities incorporated into the 
Ottoman state, and the successive frontier conquests 
in the Balkans, constituted a separate sandjak as a ter- 
ritory of command. At a rough estimate, a sandjak en- 
compassed an area of several thousand km 2 and a 
population of perhaps 100,000 on average. Usually 
reflecting pre-Ottoman administrative divisions and 
geographical realities, sandjak size and boundaries re- 
mained fairly stable through the centuries; provinces 
in modern Turkey, especially in western and northern 
Anatolia, are very similar to 15th-century sandjak 

At least until the mid- 16th century, sandjak main- 
tained two distinct but eventually merging senses, 
military command and a provincial district: in the 
sense of command of a body of troops there were, in 
addition to the sandjak beyi of a district, sandjak beyk of 
Anatolian auxiliary troops, musellem (cavalry) andyaya 
(infantry). Even the sandjak beyis proper, i.e. of a 
district, were sometimes referred to as atlu sandjak beyi, 
i.e. cavalry commander, in their primary role as the 
commander of all the (frr&t-holders, i.e. those officers 

revenue-grants in a particular district. The sandjak beyi 
was required to maintain his own military retinue 
supported by the official khass [q. v. ] revenues allocated 
to him, the number of his retinue being commen- 
surate with the size of his kjidss. In time of mobilisa- 
tion, the sandjak beyi led his own household and the 
troops of his district to join the campaign, sometimes 
entrusted with discreet military operations such as 
reconnaissance and advance or rear guard, otherwise 
marching into batde under the command of the 
beylerbeyi [q.v.] of the province. The maritime sandjaks, 
most of them included in the kapudan pasha' s [q.v.] pro- 
vince of DjazdHr-i Bahr-i Sefid (q.v.; literally, Aegean 
Islands, but also including mainland Anatolian and 
Grecian districts) supplied ships for naval campaigns 
instead of (but sometimes as well as) cavalry troops. 
Eventually, the sense of district for sandjak (also 
liwa*, especially in documents) and district-governor 
for sandjak beyi (also mir liwa 7 ) came to predominate. 
Sandjak can be considered the main administrative 
division in the Ottoman empire in various senses. For 
one thing, in lesser dirliks the area supplying the dirlik- 
holder's income was co-extensive with the limits of his 
authority. For higher level officers, the dirlik, khass, 
was normally wholly included within the territory, but 
the territory governed was much larger than the khass. 
A sandjak beyi usually derived his khass income from the 
main towns of his district, the percentage of urban vs. 
rural taxes constituting his khass varying according to 
the level of town development and commercial taxes 
in each district; the rest of the sandjak might support 
a dozen or so officers' dirliks (ze'dmel) and a few hun- 

dred timars for cavalrymen (sipahi [see sipahi. 1]). In 
other words, it is at the sandjak level that the ad- 
ministrative unit was much larger than an officer's 
revenue source. The governor-general (beylerbeyi) was 
also the sandjak beyi of the chief district of his province. 
The primary administrative role of the sandjak is 
underscored by the fact that provincial area regula- 
tions (kdnun-ndme [q.v. ]) as well as land and population 
surveys (tahrir) were drawn up for each sandjak. 

In 1527, after the great conquests of Selim I but 
before Hungary and eastern Anatolia were fully in- 
tegrated in the realm, there were 97 sandjaks in seven 
provinces, as well as 17 Kurdish sandjaks of special 
status (Topkapi Palace Archives, D. 5246). Later in 
the 16th century some larger districts e.g. Bosna 
[q.v.], were reconstituted as provinces; there was also 
expansion both in the north-west and in the east; the 
result is that the number of provinces and districts in- 
creased to about 35 and more than 300 respectively. 
Especially in frontier regions, the tendency was to 
create smaller provinces with just a few districts; the 

number of higher-ranking officials in sensitive areas. 
Identified very closely with the command of his 
district's troops, the position of the sandjak beyi eroded 
as the military value of provincial cavalry declined in 
the 17th century. There was a relative centralisation 
of provincial authority at the province level in the 
hands of the governor-general, who very often came 
to depend on local notables for routine administration 
rather than through sandjak beyis. This process is 
already discerned at the end of the 16th century by the 
restyling of provinces as eyalets rather than beylerbeyilik. 
Nevertheless, both the dirlik-ho\d\ng provincial 
cavalry and the sandjak beyi survived a long time after, 
though in reduced circumstances. When the dirliks 
were finally abolished in 1837 the sandjak became 
simply an administrative subdivision. The mutasarrif 
the governor of a sandjak (or liwd'' or now also mutasar- 
rifllk) was henceforth a civil official, distinct from the 
mir liwa> who now became the modern general of the 
brigade. The division into sandjaks was maintained by 
the 1864 and 1871 laws of the wilayets (the former 
eyalets). The term was finally abolished in provincial 
administration by the Ankara Grand National 
Assembly in the 1921 Constitution. 

Bibliography: For older bibl., see J. Deny's EI 1 
art., to which should be added Pakahn, s.v. Sancak; 
for a summary of Ottoman provincial administra- 
tion, see H. Inalcik, The Ottoman empire. The classical 
age, 1300-1600, London 1973, ch. 13; M. Kunt, 
Sancaktan eyalete, Istanbul 1978, and idem, The 
Sultan 's servants. Transformation of Ottoman provincial 
government, 1550-1650, New York 1983, include 
discussion of lists of provinces and districts; 
Anatolian historical geography is treated in T. 
Baykara, Anadolu 'nun tarihi cografyasina girts: 
Anadolu 'nun idari taksimati, Ankara 1988; for the em- 
pire as a whole see A. Birken, Die Provinzen des 
Osmanischen Reiches, Wiesbaden 1976, and D.E. Pit- 
cher, An historical geography of the Ottoman Empire, 
Leiden 1972; for provincial regulations see sandjak 
kanin-ndmes in Barkan, Kanunlar; since M.T. 
Gokbilgin, XV. -XVI. asirlarda Edirne ve Pasa livasi, 
Istanbul 1952, and inalcik, Surel-i defter-i sancak-i Ar- 
vanid, Ankara 1954, numerous studies of individual 
Anatolian and European sandjaks have appeared. 
(J. Deny-[M. Kunt]) 
SANDJAK BEYI [see sandjak]. 
SANDJAK-I SHERIF, Liwa'-i SherIf or c Alem-i 
NebewI, in Ottoman Turkish usage the sacred 
standard of the Prophet Muhammad which is 


kept in the palace of Topkapi at Istanbul (see topkapi 
sarayi) together with the other holy relics of Islam 
(Emdndt-i Mubdreke or Mukaddese) such as the Holy 
Mantle (the Burda) [see khirka-y! sherTf], Holy 
Footstep and Beard of the Prophet (see kadam sharif 
and lihya-yi sharif]. 

According to a tradition recorded by Mouradgea 
D'Ohsson (Tableau general de I'Empire Othoman, Paris 
1788-1824, ii, 378) and the Ottoman historian 
Findiklil! Mehmed Agha (Sildhdar TarHkhi, ed. Ahm- 
ed Reffk [Altmay], Istanbul 1928, i, 14) this is the 
black banner known as c Ukdb used among other stan- 
dards in the battles against Kuraysh. It was used as a 
door curtain by c A : 'isha as well. C A1T carried it at the 
conquest of Mecca and handed it to c Amr b. al- c As 
[q.v.\ during the battle of Sifffn (37/657). During the 
reigns of the first four caliphs, this battle standard was 
always planted in front of the troops. It was handed 
down from the Umayyads to the c Abbasids and was 
handed to the Ottoman sultan Selfm I [q. v. ] by Kha'ir 
Beg [q. v. ] after the conquest of Cairo in 922/ 1 51 7 . An- 
other tradition has it that it was presented to the Ot- 
toman sultan by Abu Numayy, son (and successor) of 
the Sharif of Mecca Abu '1-Barakat, at the same time, 
as a symbolic expression of his submission. In this 
case, the colour is given as green and its size as 
0.13 x 113 cm. However, according to the historian 
Mustafa C A1I (948-1008/1541-1600) [see c ali], Sultan 
Selfm did not carry the sacred standard with him to 
Istanbul at the time. It appears to have been deposited 
in the provincial treasury of Damascus. A standard of 
the Prophet reportedly usually accompanied the year- 
ly Hadjdj caravan from there. Ewliya Celebi [q.v.] 
mentions having seen the sacred standard next to the 
Mahmal [q.v.] at the occasion of the departure of such 
a caravan in Damascus in 1672 (see S. Faroqhi, Herr- 
scher uber Mekka: die Geschichte der Pilgerfahrt, Munich- 
Zurich 1990, 52). The Ottoman historian Selaniki 
[q.v.], confirmed by C A1I, narrates that in 1001/1593 
at the outbreak of the so-called "Long War" against 
the Habsburg Emperor, it was decided to have the 
sacred standard from Damascus brought to the army 
in the field in order to raise the morale of the troops. 
The following years, each campaigning season, the 
standard travelled up and down under an escort of the 
Damascus Janissary contingent. In 1003/1595 it was 
decided to deposit the sacred standard definitively in 
the palace of Topkapi so that it could be kept together 
with the other relics of the Prophet. On 23 Shawwal 
1004/20 June 1596, Sultan Mehemmed III [q.v.] took 
the standard with him leading the army in person. At 
that occasion it was escorted by 300 sayyids led by the 
so-called sandjak-i sherif sheykhi. During the battle it was 
planted in front of the tent of the sultan, with hafizs 
reciting sura XLVIII (al-Fath). The sacred standard 
probably stood before the Sultan's person during the 
Battle of Mezo-Keresztes/Hac Owasi [q.v.], 23-6 Oc- 
tober 1596, although this is only mentioned by 
Selaniki and not by a number of eye-witness accounts 
which do mention the sultan as putting on the Holy 
Mantle and handling the sword of the caliph C A1I, 
Dhu'l-Fakar [q.v.]. 

From this time onwards, the sultans, when not join- 
ing their armies in the field, appointed the Grand 
Vizier serddr-i ekrem, commander-in-chief (see bab-i 
ser c askeri). Upon leaving the capital, the Grand 
Vizier went in a ceremonial procession to the imperial 
tent erected in camp at Dawud Pasha (in the case of 
war in Europe) or Haydar Pasha (in the case of war 
in Asia), there to receive the insignia of office from the 
hands of the sultan: a sable fur caftan, a Selimi turban 
with one or more aigrettes (sorguc), a gilt sword and 

the sacred standard. The sultan wished his vizier suc- 
cess and the latter left, carrying the sacred standard on 
his shoulder. Upon the return of the army, a like 
ceremonial took place. 

When the Kapuddn Pasha put out to sea, he offered 
a ceremonial salute to the Sandjak-i Sheriff exhibited for 
the occasion in the arsenal (Tersdne-yi c Amire [see ter- 
sane]) in the presence of the Grand Vizier. The sacred 
standard was also brought forward in times of 
rebellion. An exceptional occasion was Mustafa Pasha 
Bayrakdar's [q.v.] planting it in front of his troops 
entering the capital to depose Sultan Mustafa IV 
[q.v.] on 4 Djumada II 1223/28 July 1808 (General 
State Archives The Hague (ARA), Van Dedem 
Papers 2.21.049-61). 

Six Grand Viziers were killed in action while the 
sacred standard was with them, including Khadim 
Sinan Pasha on 29 Dhu '1-Hidjdja 922/1517 while 
defending the person of the sultan against rebel 
soldiery, and Shehid c Ali Pasha on 16 Sha c ban 
1 1 28/ 1 7 1 5 with the sacred standard in his hands in the 
Battle of Grosswardein/Nagyvarad, leading the 
counter attack against the imperial army. Count Mar- 
sigli, describing the various banners and standards in 
use with the Ottoman army, mentions the standard of 
the Prophet. He never saw it deployed either in camp 
or on the march. He concludes from the event that the 
Ottomans were always successful in saving the sacred 
standard from falling into the hands of the enemy in 
their many defeats during the war in Hungary (1683- 
99) because it was always heavily escorted. After the 
disastrous battle of Salankamen (24 Dhu '1-Ka c da 
1 102/19 August 1691), the escort was richly rewarded 
for bringing it safely back home. According to the 
story, a miracle happened which made the banner in- 
visible when passing through the enemy cavalry. 
(L.F. Marsigli, Stato militare dell'Imperio Ottomanno, 
The Hague 1732, repr. Graz 1972, ii, 51-2). 

By the end of the 1 lth/1 7th century, the sacred 
standard was badly worn. The original black banner 
was replaced by three green silk banners, to each of 
which were attached pieces of the Sandjak-i Sherif, thus 
transferring its blessed powers to the new standards. 
One of these was handed to the outgoing 
commanders-in-chief, one the sultan kept with his 
own person when travelling outside the capital and 
one was permanently kept in the treasury of the 
Topkapi Palace (in the Khirka-yi Sherif DaHresi). The 
lasting respect in which the sacred standard was held 
by the Ottomans is exemplified by the traditional view 
that all men between seven and seventy years of age 
were obliged to join the djihdd [q.v.] when the standard 
was brought forward. The place in front of the throne 
room in the Topkapi Palace where it used to be 
planted was held to be sacrosanct and not to be trod- 
den by anyone's feet. Till 1908, two soldiers with fix- 
ed bayonets stood on guard there. 

The handling of the sacred standard was regulated. 
Texts of such regulations (kdnun-ndme) of different 
periods are extant. When Sultan Mahmud II [q.v.] 
ordered the destruction of the Janissaries (see yeni 
<5eri) on 9 Dhu '1-Ka c da 1241/15 June 1826, the loyal 
"people of Muhammad" were called to- gather 
around the sacred standard, which stood planted upori 
the minbar of the mosque of Sultan Ahmed I. The last 
time the Sandjak-i Sherif was deployed was on 25 Dhu 
'1-Hi djdj a 1332/14 November 1914 at the proclama- 
tion of the so-called "Holy War" (Djihdd-i Ekber) 
against the Entente Powers. 

Bibliography: i.H. Uzuncarsih, Osmanli 

devletinin saray leskildti, Ankara 1945 repr. 1984, 241, 

244, 248-260; Pakalin, iii, 113-16; TewkM 


'Abdurrahman Pasha, Kanunndme, in MTM, i/3 
(1331/1912-13), 497-544; Selaniki Mustafa Efendi, 
Tanh-i Selaniki, ed. M. ipsirli, 2 vols., Istanbul 
1989, i, 322, 420-1, ii, 493, 534, 608, 611; Miihur- 
dar Hasan Agha, Diewahir ul-tewarikh, tr. E. Pro- 
kosch, Krieg uniSieg in Ungarn, Graz, etc. 1976, 24- 
5; M. Cezar, M. Sertoglu et alii, Mufassal Osmanli 
tarihi, ii, Istanbul 1958, 767-8; K. gig, Islamische 
Reliquien, Istanbul 1966; J. Schmidt, The Egri cam- 
paign of 1596. Military history and the problem of sources, 
in A. Tietze (ed.), Habsburgisch-osmanische Bezie- 
hungen. Ciepo Colloque Wien 1983, Vienna 1985, 
125-44. (A.H. de Groot) 

SANDJAR b. Malik ShAh, c Adud al-Dawla Abu 
'1-Harith Ahmad, Saldjuk malik in Khurasan 490- 
511/1097-1118 and then supreme sultan of the 
Great Saldjuks, ruling Khurasan and northern Per- 
sia till his death in 552/1 157; he accordingly ruled for 
some 60 years. The name Sandjar, which occurs for 
other members of the Saldjuk family and elsewhere in 
the Turkish world, seems to mean in Turkish "he 
who pierces, thrusts", cf. M.Th. Houtsma, Em 
turkisch-arabisches Glossar, Leiden 1894, text 29, 
glossary 78, 80, and the detailed discussion by P. 
Pelliot, in Oeuvres posthumes, ii, Paris 1949, 176-80; a 
contemporary European rendering is given by Otto of 
Freising in his chronicle, sub anno 1145: Saniar- 
dosl Samiardos fratres. 

He was born in either Radjab 477/November 1084 
or Radjab 479/November 1086, his mother being one 
of Malik Shah's concubines. Whilst still a boy, he was 
in 490/1097 appointed by his half-brother Berk-yaruk 
[q.v.] as governor of Khurasan after the unsuccessful 
revolt there and death of Arslan Arghun b. Alp 
Arslan. During the internecine struggles over the 
supreme sultanate between Berk-yaruk and Muham- 
mad b. Malik Shah [q.v.], Sandjar generally took the 
side of his full brother Muhammad, but from the con- 
stitutional aspect regarded himself as governor only of 
the eastern provinces and as subordinate to the 
supreme sultan in the western lands, calling himself 
on his coins of this time merely a malik and 
acknowledging Berk-yaruk and then Muhammad as 
al-Sultan al-Mu c azzam. 

However, when Muhammad died in 511/1118, 
Sandjar refused to consider himself subordinate to his 
nephew in the west, Mahmud b. Muhammad [q.v.], 
and as the senior member of the Saldjuk family, both 
his de facto power and his position under Turkish tribal 
custom gave him a claim to the supreme sultanate 
even though this had previously been held, for eighty 
years, by the Saldjuk who controlled western Persia 
and 'Irak. The squabbling sons of Muhammad b. 
Malik Shah were too divided and militarily weak to 
dispute Sandjar's position, and they had generally to 
place Sandjar's name plus his title of al-sultan al- 
mu'-azzam on their coins before their own names and 
titles. The only serious opposition at the outset to San- 
djar's claims here came from Mahmud, but in 
513/11 19 Sandjar marched westwards with a powerful 
army, whose commanders included, besides the 
sultan himself, four vassal kings, defeated Mahmud at 
Sawa [q.v.] in northern Djibal and marched onwards 
to Baghdad. When peace was made, Mahmud agreed 
to Sandjar's supremacy and was made the latter's heir 
(in the event, he died long before Sandjar did), but 
had to relinquish to Sandjar the Caspian provinces of 
Mazandaran and Kumis and the town of Rayy, the 
key point for control of northern Persia, and to agree 
to the re-appointment of Sandjar's shihna or military 
governor in Baghdad. It was during these year also 
that Sandjar was again concerned with the Isma'Tlis of 

northern Persia and Khurasan; in 497/1 104, when he 
was malik, he had sent an expedition against these 
sectarians in Tabas, and now in 520/1126 his vizier 
Mu'in al-Dln Mukhtass al-Mulk campaigned in 
Kuhistan (an action which doubtless contributed to 
his death by assassination in the following year). In 
the 1140s, however, Sandjar and the Isma c IlI leaders 
who had succeeded to Hasan-i Sabbah [q.v.] came to 

On Mahmud's death in 525/1131, his brothers 
Mas c ud, Toghril (II) and Saldjuk Shah successfully 
disputed the succession of Mahmud's young son 
Dawud, but were unable to agree amongst themselves 
as to who should be sultan. They laid the question 
before Sandjar, as senior member of the dynasty. San- 
djar's favoured candidate was Toghril, but his preoc- 

below) prevented him from providing Toghril with 
much military support, and the latter died anyway in 
529/1 134, allowing Mas c ud to succeed in the west and 
to reign there for twenty years. Sandjar's last major 
intervention in the affairs of the family in the west had 
been his defeat of Mas c 0d at DInawar in 526/1132, 
but thereafter, affairs in Khurasan and Transoxania 
increasingly claimed his attention. 

Sandjar continued to exercise the overlordship over 
the Karakhanids of Transoxania [see ilek-khAns] first 
imposed Dy his father Malik Shah, but had on various 
occasions to lead expeditions acrosse the Oxus against 
recalcitrant Khans. In 495/1102 he had stemmed at 
Tirmidh the invasion of one Karakhanid claimant 
and had placed on the throne in Samarkand Arslan 
Khan Muhammad (II). But towards the end of the 
latter's long reign, in 524/1130, Sandjar had to come 
with an army to reinforce the Khan's faltering 
authority in Samarkand. He set up various 
Karakhanid nominees on the throne there, ending 
with (possibly in 526/1132) Arslan Khan Muham- 
mad's third son Mahmud, who was Sandjar's 
nephew, since his mother Terken Khatun was San- 
djar's sister. The fortunes of Sandjar and Mahmud 
were to be closely interwoven over the ensuing year; 
when Sandjar was captured by the Oghuz in 548/1 1 53 
(see below), Mahmud was recognised by the Saldjuk 
army in Khurasan as interim sultan of Khurasan and, 
after Sandjar's death, likewise as legitimate ruler 
there till his own death in 557/1162. 

As ruler of Khurasan, Sandjar was also concerned 
with the neighbouring great power to his east, the 
Ghaznawids. Since the peace agreement between 
Caghri Beg and the Ghaznawid Ibrahim b. Mas'Qd in 
451/1059, there had been a considerable Saldjuk 
cultural penetration of the Ghaznawid empire, visible 
for instance in numismatic patterns, titulature and 
literary trends. A raid early in his reign by Sandjar in- 
to Ghur [q.v.], the mountainous central region of 
Afghanistan, is recorded, but a succession dispute 
within the Ghaznawid royal house also allowed San- 
djar to extend direct Saldjuk sovereignty over the 
Ghaznawid lands of eastern Afghanistan. When 
Arslan Shah came to the throne of Ghazna in 
509/1115, his brother Bahram Shah escaped to 
Khurasan and appealed to Sandjar for help. The 
Saldjuk sultan marched eastwards with a formidable 
army, defeated Arslan Shah, sacked Ghazna and 
placed Bahram Shah on the throne there (510/1117). 
The latter agreed to become Sandjar's vassal, to place 
his name first in the khutba and on the coinage and to 
pay an annual tribute of 250,000 dinars. For some 
thirty years, Bahram Shah acknowledged this subor- 
dinate status, only once becoming restive when in 
529/1 135 Sandjar and his other vassal, the Kh w arazm 

Shah Atsiz, came with their forces to Ghazna, expell- 
ing Bahrain Shah to India before he returned and 
agreed to resume his vassalage. 

Along the northern fringes of Khurasan, Sandjar 
had as vassals, in addition to the Karakhanids, the 
Kh w arazm Shahs of the line established by his father 
Malik Shah, sc. the line of AnQshtigin Qharca 5 ! [see 
kh w arazm-s_hahs]. The second ruler of this line, 
Kutb al-DIn Muhammad, was Sandjar's faithful 
vassal, as was initially his son and successor '■Ala 3 al- 
DTn Atsiz, attending Sandjar's court regularly. But 
relations deteriorated as Atsiz built up his own 
military strength and began to identify himself with 
the particular interests of his province, until in 
533/1138 he rebelled openly. Sandjar invaded 
Kh w arazm, drove out Atsiz and left there a Saldjuk 
prince and his atabeg; but as on earlier occasions 
when outside powers had endeavoured to impose their 
rule over the province, the Kh w arazmian people rose 
up against the occupiers and expelled them, enabling 
Atsiz to return and take the offensive against Trans- 
There now appeared a new force in the affairs of 
Central Asia, the Kara Khitay [q.v.\, the K'i-tan or 
Liao of the Chinese sources. Within the western 
Karakhanid principality, disaffected Karluk [q.v.] 
tribesmen called in the Kara Khitay; Mahmud Khan 
of Samarkand in turn appealed to his suzerain and 
kinsman Sandjar. The latter appeared in Transoxania 
with a large army, but was in 536/1141 defeated by 
the Kara Khitay in a bloody battle on the Katwan 
steppe of UshrQsana [q.v.], on the middle Syr Darya. 
Sandjar and Mahmud Khan fled to Khurasan, aban- 
doning Transoxania, and the Kara Khitay went on to 
make Atsiz their own vassal; accordingly, whilst San- 
djar's defeat was clearly opportune for Atsiz, it seems 
improbable that the Shah had, as some of the Islamic 
sources assert, incited the Kara Khitay to invade as an 
act of revenge on Sandjar for the sultan's killing of his 
son Atligh. At this point, Atsiz raided into Khurasan 
himself, but was driven back by a Saldjuk counter- 
invasion of Kh w arazm which penetrated to the capital 
Gurgandj and compelled the Shah to disgorge the 
treasuries which he had previously looted from San- 
djar's capital of Marw (538/1 143-4). Yet once again, 
Kh w arazm proved too hostile for Saldjuk troops to be 
able to remain there. 

However, the tragic end to Sandjar's reign and the 
resultant downfall of Saldjuk power in the east came 
about not from the attacks of external foes like the 
Kara Khitay or from those of rebellious vassals like 
Atsiz, but from an explosion of internal discontent 
within Khurasan itself, largely brought about by the 
policies of Sandjar's aides and officials there. 
Khurasan and the steppes to the southeast of the 
Caspian Sea, in Gurgan and Dihistan, contained ex- 
tensive pasture grounds which supported numerous 
groups of tribally-organised Turkmens. These includ- 
ed Turkmens driven southwards into the Khurasa- 
nian fringes by the recent upheavals in the Central 
Asian steppes, including the pressures from the 
Kh w arazm Shahs and the Kara Khitay, and also 
descendants of the Oghuz tribesmen whose dynamic 
had brought the Saldjuks to power in the Islamic lands 
a century earlier. The sultans had accordingly always 
felt certain obligations towards these kinsfolk of theirs, 
often making special administrative arrangements for 
them in the regions where they were especially 
numerous, appointing special officials (shihna, ra?is 
[q.vv.]) to act as channels of communication between 
the nomads and the Saldjuk state, whose ddminating 
Perso-Islamic ethos was now largely alien to the 

Turkmens. These arrangements now came under 
severe strain because of the financial exigencies aris- 
ing from Sandjar's military adventures, increasingly 
expensive after 529/1135; he is said to have disbursed 
three million dinars for the Transoxanian campaign of 
536/1141, not counting the cost of presents and robes 
of honour given to various local potentates. The 
burden of taxation in order to pay for these fell on 
sedentaries and nomads alike, but the Oghuz in the 
upper Oxus regions of Khuttal and Tukharistan 
[q. vv. ] finally rebelled against the tax demands and the 
harsh collecting methods of the shihna over the 
Turkmens there, the slave commander c Imad al-DIn 
Kumac of Balkh. Despite placatory approaches from 
the Oghuz, Sandjar insisted on mounting punitive ex- 
peditions against them, but he was twice defeated, 
forced to evacuate his capital Marw and finally cap- 
tured by the nomads (548/1153). The Oghuz bands 
swept through Khurasan, attacking towns there and 
showing particular violence and hostility towards 
members of the Saldjuk administration and the 
religious classes, closely linked with the Saldjuk state; 
a general climate of insecurity was created in both 
towns and countryside of Khurasan, in which various 
other anti-social elements such as the '■ayyars [q.v.] 
took advantage of the breakdown of authority to fur- 
ther their own interests. 

The leaderless Saldjuk army in Khurasan offered 
the throne there to the refugee Karakhanid Mahmud 
Khan who, as the son of Sandjar's sister, had Saldjuk 
blood in his veins, and the Saldjuk sultan in the west, 
Muhammad (II) b. Mahmud [q. v. ], agreed to this and 
sent an investiture diploma. In fact, over the next few 
years, real power in Khurasan fell into the hands of 
Saldjuk amirs, such as Mu'ayyid al-DIn Ay Aba at 
NIshapur and Ikhtiyar al-DIn Ay Tak at Rayy, for 
Mahmud Khan was never able to establish firm con- 
trol over the whole of Khurasan, and he died in any 
case by 559/1164. Sandjar, meanwhile, was carried 
round by his Oghuz captors for three years, apparent- 
ly in humiliating circumstances and enduring hunger 
and deprivation, until he managed in 551/1156 to 
escape to Tirmidh and Marw. But a year later he 
died, aged 71, and with him, the authority of the 
Saldjuks in eastern Persia ceased; to contemporaries it 
seemed like the end of an epoch. 

Sandjar's court, when he was not campaigning, was 
normally based on Marw, where he also had a fully- 
developed administration, headed by the diwan-i aHa, 
presided over by a series of viziers, of whom eight are 
known, from Shihab al-Islam c Abd al-Razzak (511- 
16/1117-22) to Nizam al-Mulk Hasan (at some point 
after 547/1152). These were usually Persians or 
Arabs, though from 516/1122 to 518/1124 Sandjar 
had a Turkish vizier, Muhammad b. Sulayman 
Kashgharl Yighan (or Toghan) Beg. We know some- 
thing also of provincial administration within San- 
djar's dominions, inter alia from the collection of ad- 
ministrative documents, the c Atabat al-kataba, made by 
Muntadjab al-DIn Badl* Djuwaynl, who was head of 
Sandjar's chancery, diwan-i insha 1 , at Marw. This ad- 
ministration was exercised through centrally ap- 
pointed officials with designations like wdli, ndHh, 
shihna and ra-'is, although this last had an additional 
role as representative of the urban notables vis-a-vis 
the sultan [see ra 3 Is. 2], and we possess from San- 
djar's reign documents on the nomination of provin- 
cial governors for Gurgan and its dependencies, 
Mazandaran, Rayy, Balkh and its dependencies, 
Marw, Tus and Dihistan. 

The court at Marw was also a centre for Persian 
literary activity, under the patronage of the sultan and 


of his great officials and commanders. Barthold's 
categorical assertion that Sandjar was illiterate 
(Turkestan, 308) requires further proof before this can 
be accepted, since we know that some at least of his 
kinsmen and contemporaries amongst the western 
Persian Saldjuks were highly literate. The Persian 
poet Mu'izzl (d. ca. 519-21/1125-7 [?.».]) was San- 
djar's chief eulogist during the earlier part of his 
reign, and it was to seek Mu'izzT's intermediacy that 
the anecdotist Nizaml c ArudI Samarkandl [<?.».] came 
to Sandjar's court when it was at Jus in 510/1116-17. 
Also, AdTb Sabir. \q.v.\ served both as Sandjar's 
panegyrist and on official missions, and it was in this 
latter function that the poet was killed by the 
Kh~arazm Shah Atsiz (in the mid-1 140s). The great 
theologian al-Ghazali [q. v. ] sent a letter to, and made 
a speech before, Sandjar (see Makdtib-i fdrsi-yi Ghazdli 
bi-ndm-i FaddHl ai-andm min rasdHl Hudjdjal al-Isldm, ed. 
c Abbas Ikbal, Tehran 1333/1954, 3-5, 6-10 (with 
interesting introd. by the original compiler), German 
tr. Dorothea Krawulsky, Bnefe und Reden des Abu Hdmid 
Muhammad al-Gazzdli, Friburg-im-Br. 1971, 63-76. 
Bibliography: 1. Sources. Bundarl, 125 11'., 
224-5, 255-84; NIshapuri, Saldjuk-nama, 44-52; 
Sadr al-Din Husayni, ed. Iqbal, 64-5, 77-8, 84-96, 
99-101, 121-2, 123-6 (Eng. tr. Qibla Ayaz, ^n unex- 
ploited source for the history oj the Saljuqs ..., Edinburgh 
Univ. diss. 1985, unpubl.); RawandT, 167-203; 
Yazdl, <Urdda, 94-113; Ibn al-Athlr, x-xi; Ibn 
Khallikan, ed. c Abbas, ii, 427-8, tr. de Slane, i, 
600-2; Djuwaynl-Boyle, i, 278-88 (on the relations 
of Sandjar and Atsiz). 

2. Studies. W. Barthold, Turkestan down to the 
Mongol invasion 3 , 319-32; M.A. Kdymen, Bityuk 
Selcuklular imparatorlugunda Oguz isyani, and Buyuk 
Selcuklu imparatorlugu tarihmde Oguz ishlasi, in AUDT- 
CFDergisi, v (1947), 159-73, 563-620, with German 
tr. at 175-86, 621-60; idem, Buyuk Selcuklu im- 
paratorlugu tarihi. II. Ikinci imparatorluk deon , Ankara 
1954; idem, Selcuklu devn tilrk tarihi, Ankara 1963; 
idem, 1A art. Sencer; M.G.S. Hodgson, The order of 
Assassins, the struggle of the early Nizdri IsmaHlis against 
the Islamic world. The Hague 1955, 88, 100-2. 
146 ff.; A.K.S. Lambton, The administration of San- 
jar's empire as illustrated in the c Atabat al-kataba, in 
BSOAS, xx (1957), 367-88; H. Horst, Die Staatsver- 
wallung der Grosselguqen und Horazmsdhs (1038-1231), 
Wiesbaden 1964; C.E. Bosworth, in Camb. hist. 
Iran, v, 108, 110, 112-13, 119-20, 135-59; Lamb- 
ton, in ibid, ch. 2, index; N.M. Lowick, Seljuq 
coins, in NC, 7th ser., vol. x (1970), 245-50 (also in 
Coinage and history of the Islamic world, Variorum Edi- 
tions, London 1990, no. VI). 

(C.E. Bosworth) 
SANF, a geographical term appearing in the ac- 
counts of Arab travellers from the mid-8th century 
and denoting at times an island, and at others a 
kingdom of the mainland, bordering on the 
sea, or a sea. Study of these itineraries makes it clear 
that they refer to Campa or Champa situated between 
Cambodia and the delta of the Song Cot in Viet Nam. 
The information given by these authors is very 
laconic, and, curiously, the 13th and 14th century 
texts are much less well documented than the earliest 
ones. There is practically nothing on the people there, 
sometimes described as being brown, nor on the 
political situation, except that Ibn al-Nadim \q.v.) 
(end of the 10th century) speaks of a victorious war by 
the Viet Nam which ravaged the land. All accounts 
stress the production there of aloes wood, considered 
as excellent, except for Yakut in his Mu'-djam al-buldan 
(end of the 12th-opening of the 13th century), who 

considered this product as the worst possible. Al-IdrisT 
(12th century) is the fullest, speaking of stock-rearing 
(bovines and buffaloes) and agriculture (rice, sugar 
cane, coconut palms and bananas). He describes 
Hindu-type customs (respect for cows), whereas Ibn 
al-Nadim merely mentions Buddhist temples, which 
can only have been rare at his time. These texts also 
contain tall stories, such as that in the Mukhtasar al- 
< -adj_a : 'ib (beginning of the 1 1th century) on the legend 
of the island of the White Palace, made of crystal, 
which Alexander the Great allegedly visited. 

Sanf seems to have been a stage much frequented 
by Muslim merchants travelling between the Indian 
Ocean and the China Seas. Better known under its 
Sanskrit name of Campa, the plains around the deltas 
were inhabited by populations of controversial origin 
but who belonged to the Austronesian linguistic fami- 

(Sedang, Bahnar, Jorai, Rhade, etc.), speaking 
Austronesian or Mon-Khmer languages, were more 
or less in a vassal status. 

The ancient Cham seem to have been organised in 
principalities which were at first independent, stretch- 
ed along between the Hoanhso'n and the Phan-rang 
region. The first to appear in history was the most 
northerly one, founded in A.D. 192 and known under 
the Chinese name of Lfnyi. Before the beginning of 
the 7th century, all these principalities must have been 
brought together. Campa had a very lively history. 

sailors, willingly turning to piracy. The land was torn 
apart by frequent revolts and succession wars. As an 
agressive neighbour, it often fought with Cambodia, 
occupied Angkor 1177-81, but became a Khmer pro- 
vince 1202-20. During 1283-5, it successfully repelled 
a Mongol invasion. But the main enemy was Viet 
Nam. Having the advantage until the 10th century, it 
then fell behind, alternately suffering defeats and 
making counter-attacks. The king Bong Nga (1360- 
90) invaded the delta of the Song Coi three times. 
After him, decadence set in. In 1471 the Vietnamese 
emperor Le Thanh Ton crushed the kingdom, which 
became a vassal state, gradually nibbled away and 
progressively depopulated. There remained only a 
few enclaves and a shadow of former royalty. In 1832, 
the last king and part of the population took refuge in 

Inspired by Indian culture, the Cham created a 
brilliant and varied civilisation whose architectural re- 
mains, for long neglected, are now being restored. 
Society was in origin matrilineally organised, and it 
adopted the Hindu system of castes and customs. At 
the present time, there remain few Cham in Viet 
Nam, most of them preserving an impoverished form 

The presence of colonies of Muslim merchants is 
known from the middle of the 10th century, and a 
Muslim even led two embassies to China in 958 and 
960 (see Hudud al-<dlam, tr. Minorsky, tr. 86, comm. 
240). It is unknown when Islam penetrated Cham 
society, but this was probably at the end of the 15th 
century. Yet it has always remained a minority faith, 
and one varying a great deal from the fundamental 
norms [see further Cam]. 

Bibliography: Jeanne Luba, Un royaume disparu, 
les Cham et leur art, Paris 1923; G. Maspero, Le 
royaume de Champa, Paris 1928; J.-Y. Claeys, 
Simhapura, la grande capitale chame ( VI'- VIII' s. A.D. ), 
in Revue des Arts Antiques, vii (1931), 93-104; J. 
Boisselier, La statuaire du Champa, Paris 1963; G.R. 
Tibbetts, A study of the Arabic texts containing material 
on South-East Asia, Leiden 1979; Po Dharma, Le 

Pan(luranga (CampS), 1822-1832, ses rapports avec le 

Vietnam, diss. Paris 1987. See also the bibl. to Cam. 
(P.L. Lamant) 

SANHADJA, an important group of Berber 
tribes who played an historical role in North Africa 
from the 4th/10th century onwards. They lived in the 
two Maghribs and in Ifrfkyia; some were sedentary, 
whilst others had moved into the desert and become 
nomadic. According to the Berber genealogists, the 
Sanhadja were one of the seven great tribes descended 
from Bernes, son of Berr. On the other hand, for the 
Arab genealogists, like Ibn al-Kalbl, they had, in 
common with the Kutama or Ketama, a Yemeni 
origin, and had allegedly been sent to the Ma gh rib by 
Ifricos, one of the kings of Yemen. 

The whole history of the Sanhadja was dominated 
by their opposition to another great group of Berber 
tribes, the Zanata, in particular, to the Miknasa, 
Maghrawa [q.v.] and the Banu Ifran \q.v.]. According 
to Ibn Khaldun, the Sanhadja had almost 70 
branches, one of the most important being the 
Talkata who occupied part of the central Ma gh rib. 
The first known chief of the Talkata was probably 
Manad b. Mankus, whose son Ziri was the ancestor 
of the ZTrid dynasty [q.v.] and the founder of their 
capital, Ashlr, in the Djabal Titteri (324/935). 

As partisans of the Fatimids, this branch of the 
Sanhadja had adopted ShiSsm (they were later, how- 
ever, to return to Sunnism), and when the Fatimid 
caliph al-Mu c izz [q.v.] left for Egypt (361/972), he en- 
trusted the government of Ifrlkiya to the ZIrids. Since 
they then became practically independent, the ZIrids 
ruled in Ifrlkiya and the eastern part of the central 
Ma gh rib until the opening of the 12th century A.D. 
(498/1105). Their break with the Fatimids in Cairo 
provoked the unleashing against Ifrlkiya by the 
Fatimids of the Banu Hilal [q.v.] Arabs, who 
devastated the land. Another Sanhadja kingdom had 
seen the light in the central Ma gh rib: a grandson of 
Ziri b. Manad, Hammad b. Buluggln, after having 
built the Kal c at Ban! Hammad [q.v. and hammadids], 
separated from his ZIrid cousins. In order to get away 
from the Hilalian pillagers, one of Hammad's suc- 
cessors founded Bidjaya [q.v.] (Bougie) (before 
461/1068-9). But gradually, these Sanhadja of the 
eastern Maghrib became dominated by the Hilal and 
other Arab tribes. 

At the end of the 4th/beginning of the 1 1th century, 
a son of Ziri b. Manad, Zawl, departed for Spain, 
where, after having entered the service of al-Mansur 
Ibn Abi c Amir, the regent for the Umayyad empire, 
he founded an independent state around Granada. 
This Sanhadja branch, however, abandoned al- 
Andalus in 416/1025 [see zirids of Spain]. 

Within the Sahara, there nomadised a second 
group of the $anhadja, the Lamtuna [q.v.], wearers of 
the litham or veil, the Almoravids [see al-murabitun]. 
After having conquered Morocco, these last seized 
Tlemcen, Oran and all the coastal region up to Algiers 
(475/1082). They clashed with their fellow-tribal 
brethren of the Banu Hammad and then embarked on 
the conquest of al-Andalus. 

The two Sanhadja kingdoms, descendants of Ziri b. 
Manad, in the Maghrib and in Ifrlkiya, disappeared 
around the middle of the 6th/12th century when a fur- 
ther wave of Berbers, the Almohads [see al- 
muwahhidun], conquered the whole of Barbary as far 
as Tunis. As for the Almoravid Sanhadja, they were 
crushed by the Almohads in Spain as well as in the 
Maghrib. A fraction of them, the Banu Ghaniya 
[q.v.], ruled in the Balearic Islands in the 6th/ 12th 
century. They even managed to seize Bidjaya 

(581/1185), Algiers, the Kal c at BanI Hammad and 
Gafsa, invaded the Djerid and reached Tripoli and 
Tunis in 599/1203; but they ended up being defeated 
by the Almohads (621/1224), and Sanhadja domina- 
tion of Barbary was ended. 

A few tribes of the Sanhadja lived in the Maghrib 
al-Aksa, in the Sus and in the Atlas, sc. the Lamta 
[q.v.] and the Gazzula, nomads, and the sedentary 
Haskura. Other less important groups lived, and still 
live, in the Atlantic coastal plains of Morocco, the 
Shawiya [q.v.] and the Dukkala, and above all, in the 
north, near the RIf [q.v.], such as the Bottuya and the 
Banu Uryaghul; but none of these Sanhadja tribes en- 
joyed any political power. 

Bibliography: Ibn Hawkal, Sdrat al-ard; Bakri, 
Masdlik, ed. and tr. de Slane, repr. Paris 1965; Ibn 
Khaldun, Hbar = Histoire des Berberes, i-ii, tr. de 
Slane, repr. Paris 1969; L. Golvin, Le Magrib central 
a I'epoque des Zirides. Recherches d'archeologie et 
d'histoire, Paris 1957; H.R. Idris, La Berberie orientale 
sous les Zirides X'-XW siecles, Paris 1959; V. 
Lagardere, Les Almoravides jusqu'au regne de Yusufibn 
Tasfin (1039-1106), Paris 1989 

(Chantal de La Veronne) 
al-SANHURI, c Abd al-Razzak, b. 11 August 
1895 in Alexandria, Egypt, d. 1971 in Cairo, is the 
Arab world's most distinguished scholar of 
modern jurisprudence, with the regeneration of 
Islamic law figuring prominently in his work. He is 
renowned for drafting new civil codes for Egypt, 
c Irak, Syria and Libya which incorporate Islamic legal 
rules to the extent which he considered appropriate for 
each country, and for comparative treatises on civil 
law and the sources of legal right in Islamic 

Al-Sanhuri obtained a licence en droit in 1917 from 
the Khedival School of Law (Cairo), graduating top in 
his class. He began his juristic career as wakll niydba 
in the Mixed Courts of al-Mansura and was a teacher 
in the school for SharFa judges. In 1921 he went to 
France for doctoral studies at the University of Lyon 
with E. Lambert. There he wrote two theses, one on 
English law which won the prix de these. The other, Le 
Calif at (Paris 1926) is a study of doctrine and history 
of the caliphate which concludes with proposals to 
revive that institution (abolished by the Turks in 
1924) and to reform the legal systems of Arab states. 
Returning to Cairo in 1926, al-Sanhurl was ap- 
pointed in the Faculty of Law at the National (later 
Cairo) University, and began to write treatises on the 
theory of obligations and contracts, notably c Akd al- 
tdjar (1930) and Nafariyyat al- c akd (1934). Drawing 
from comparative and Islamic law and case material 
of the Egyptian courts, he sought to make Egyptian 
and Islamic law part of comparative jurisprudence. 

In 1935 he was invited to Baghdad as professor and 
Dean of the newly-established Law School. He taught 
comparative (Islamic and Western) law and began 
work on constructing 'Irak's modern civil code. The 
technique he used to compare codifications of Islamic 
law (Madjalla and Murshid al-hayrdn) with legal rules in 
European codes is explained in Madjallat al-Kada} 
(Baghdad) (1936/2). 

Back in Egypt in 1936, he was-briefly Dean of the 
Law Faculty of Cairo University and began work on 
revising the Egyptian civil law. His views about revi-_ 
sion appear in al-Kdnun wa 'l-iktisdd (1936/6) and al-' 
Kitdb al-dhahabi li 'l-mahdkim at-ahliyya (Cairo 1938). 
Al-Sanhuri defended the final version before Parlia- 
ment in 1948 against critics who wanted a completely 
Islamic code. He claimed the revised code included all 
the Islamic law it was then possible to adopt, "having 


regard to sound principles of modern legislation." 
Al-Sanhuri was active in Egyptian nationalist 
politics and became prominent in public life from the 
1930s. He was among the founders of the Sa'dist par- 
ty (1937), Deputy Minister of Justice (1944) and 
Minister of Education (1945-6 and 1947). The pin- 
nacle of his public career came in 1949 when he was 
appointed Chief Justice of the Majlis al-dawla. He 
made it "a towering fortress of the protection of rights 
and the guardian of liberties" (Mursf, 1980). 

At first he supported the Revolution of 1952 and the 
Free Officers. Later came his concern to re-establish 
constitutional government and civilian rule. During 
the political turbulence of March 1954, al-Sanhuri 
was physically attacked at the Madjlis by demon- 
strators and never again held public office. 

Thereafter, he worked on his two major treatises: 
Masadir al-hakk fi 'l-fikh al-Islami, 6 pts. in 2 vols., 
Cairo 1954-9, and al-Wasit fi sharh al-kanun al-madani 
al-djadid, 10 pts. in 12 vols., Cairo 1952-70, completed 
the year before he died. Both are still in print and 
serve as basic reference works. 

Bibliography: Y.L. de Bellefonds, 'Abd al- 
Razzak al-Sanhuri: Masadir al-haqq fil-fiqh al-islami, in 
Revue internationale de droit compare (1958), x, 476-9; 
F. Ziadeh, Lawyers, the rule of law and liberalism in 
modern Egypt, Stanford 1968, 117-18, 137 ff., 156n., 
158n.; D. Khat(ab, Abd al-Razzak al-Sanhun: 1895- 
1971, in Madjallat al-Kada? (Baghdad), xxvi (1971); 
A. al-Djamr c . '-Abd al-Razzak al-Sanhuri, in ibid., 
xxvii (1972); A. Mursi, Ustadh al-asatidha, in al-'ld 
al-mPawi U-kulliyySt al-hukuk, Cairo 1980; F. Castro, 
'Abd al-Razzak Ahmad al-Sanhun (1895-1971): primi 

Gabrielli nel suo ottantesimo compleanno, Rome 1984, 
173-210; M. c Allam, al-Sanhiri, in Madjma'un fi 
khamsin 'am m , Cairo 1986, 158-60; E. Hill, al- 
Sanhuri and Islamic law, in Arab Law Quarterly (1988), 
33-64, 182-218; N. al-Sanhuri and T. al-Shawi, 
'Abd al-Razzak al-Sanhuri. Awrak al-shakhsiyya, Cairo 
1988; eidem, Fikh al-khilafa wa-iatawwaruha ( = tr. of 
Le Califat, with notes), Cairo 1989. 

(Enid Hill) 
al-SANHURI, Abu 'l-Hasan c Ali b. c Abd AllAh 
al-Nutubisi al-Malikl, Nur al-Dln, Egyptian usuli 
fakih and grammarian (ca. 814-89/ra. 1411-84). 

He was born at Nutubis, and he lived at Sanhur 
near Alexandria, where he learnt the Kur'an, and 
finally settled at Cairo, in the Azhar. Amongst his 
masters there were some of the most celebrated 
religious lawyers of the time, with whom he studied, 
amongst other works, the 'Akila and the Hirz al-ma'ani 
of al-Shatibl [q. v. ] . The Alfiyya of Ibn Malik [q. v. ] and 
the two Mukhtasars of Abu c Amr Ibn al-Hadjib [q.v.]. 
He also followed courses in mathematics and the divi- 
sion of inheritances. He then made the Pilgrimage, 
and spent some time at Mecca. 

He wrote a sharh on the Mukhtasar of Khalfl b. Ishak 
[q.v.], an unfinished work; a ta'lik to the Talkin of 
c Abd al-Wahhab al-Baghdadi (d. 422/1031); and two 
commentaries on the grammatical work, the Adjur- 
rumiyya of Ibn Adjurrum [q.v.] (see for the mss. of this, 
Brockelmann, II 2 , 238 no. 5 and S II, 333 no. 5; he 
is not to be confused with another commentator on the 
Adjurrumiyya, c AHb. Hasan al-Sanhuri al-Shafi c i). He 
taught at the Barkukiyya and Ashrafiyya madrasas, 
becoming in the end shaykh al-Malikiyya. His pupils in- 
cluded the famous North African fakih and mystic 
Ahmad Zarruk, as well as other outstanding Egyptian 
scholars (see a story on their numbers in Makhluf) 
who studied fikh and Arabic grammar and philology 
with him. Concerning one of them, we know that he 

studied with al-Sanhuri the Mukhtasar, part of 
Sahnun's Mudawwana and the Tafri' of Ibn al-Djallab, 
Ibn Abl Zayd's Risala, Ibn 'Askar's 'Umda, as well as 
the Irshdd(see al-Karafl, Tawshih, 184 no. 183). 

Bibliography: Sakhawl, Daw>, v, 249-51 no. 
843; KarafT, Tawshih al-dibadi, ed. A. al-ShatyawI, 
Beirut 1983, 130-2 no. 127; Ibn al-Kadi, Durrat al- 
hidjal fi asmd? al-ridjal, ii, 444; Ahmad Baba al- 
Timbukti, Nayl al-ibtihddj_, Beirut n.d., 208-9; 
Isma c Il al-Baghdadi, Hadiyyat al-'arifin, Istanbul 
1951, I, 737; M. Makhluf, Shadjarat al-nur al-zakiyya 
fi tabakat al-mdlikiyya, Cairo 1950-2, 258 no. 939; 
Kahhala, Mu>allifin, vii, 138; Zirikli, A'lam, iv, 307. 
For the intellectual context, see C.F. Petry, The 
civilian elite of Cairo in the later Middle Ages, Princeton 
1981; J. Berkey, The transmission of knowledge in 
medeval Cairo. A social history of Islamic education, 
Princeton 1992. (Maribel Fierro) 

al-SANHURI, ABU 'l-NADIA [see salim b. 

SANIYA [see na'ura]. 

SANTUR, a musical instrument still surviving 
today in the urban music-making of Turkey (very 
rare), of c Irak (integrated with the calghi quartet ac- 
companying al-makam al-Hraki [see makam], of Persia 
and of Kashmir (played solo). It is a cithara- 
tympanon-dulcimer made of wood, in the shape of a 
flat isosceles trapezoid provided with 72 to 96 strings 
of metal stretched from the string-holders (left-hand 
side of the trapezoid) to the pegs (right-hand side), 
grouped into groups of four strings and then resting 
on from 18 to 24 bridges. It is played with two wooden 
sticks (madrab) covered with tow or cotton and held by 
the musician between thumb and index finger (the 
techniques differ according to various schools); the 
strings give out a crystal-like sound prolonged by 
lengthy reverberation. 

The santur of Persia spans four octaves and 33 
degrees in heptatonic diatonic scales. Since the 
bridges are arranged in two groups of nine each, the 
ninth bridge allows the division of a note (e.g. fa 
natural and fa sharp in order to change gusha, melodic 
model). It is therefore more a question of the soloist's 
virtuosity than of modulation (talwin). 

The santur of c Irak spans three octaves and a third, 
and its two groups of twelve bridges define 48 degrees. 
The various ways of tuning the strings provides in the 
same register the possibility of modulating between 
genres (djins) and of passing from a Bayati (A-B<l-C- 
D I,) to a Saba (A-Bj-CH-D I,) and a Hidjaz (A-Bt- 
CH-D ([).* It thus leads more to modulation than to 

The origins of the santur are unclear. It has been 
alleged to be Assyrian(?). It does not appear in 
classical miniature paintings; it may be evoked in the 
K. al-Adwar of Safi al-DIn al-Urmawi [q.v.] (Baghdad, 
7th/13th century); it appears on a fresco of the Cihil 
Sutun palace in Isfahan (llth/17th century); it is 
mentioned in the Comte de Gobineau's Trois ans en 
Asie, Paris 1859; and it appears at the end of the 19th 
century on the "chromolithographs" of Kadjar- 
period Persia. At the beginning of the 20th century, 
it was very common in the coffee houses of Istanbul, 
Tehran and Ba gh dad, and took on a connotation of 
being connected with amusement and play, before il- 
lustrating the type of music called "traditional" by 
the West. Its present-day most brilliant exponent is 
the Persian Faramarz Pgyvar. 

* Musical abbreviations: i = less one or two com- 
mas; H = sharpened by a lemma. 

Bibliography: Faramarz Payvar, Dastur-i santur 

(method), Tehran; idem, Si kaf-i caharmidrab bara-yi 


virtuosity for the santur), 
Tehran, both publ. after 1960; The new Grove dic- 
tionary of musical instruments, London 1984, iii, 291-2 
(good bibl.). (J.-C. Chabrier) 

SANTURIN ADASI, the Turkish name for San- 
torin, Grk. Thera, Lat. Sancta Irini, the volcanic 
island which is the southernmost of the Aegean 
Cyclades group, to the north of Crete. 

It may already have suffered from Arab raids from 
Crete in the 3rd/9th century, from the Arabs of Sicily 
and then from Western corsairs, although it is record- 
ed as inhabited in ca. 549/1154 by al-ldrlsl (tr. 
Jaubert, ii, 127), the first to employ the name San- 
turin (< the island's patroness St. Irene) (see A. Sav- 
vides, Notes on mediaeval Theral Santorin until the late 15th 
century, in Pariana, xv, no. 53 [1994], in Greek). 
Turkmen raids from Menteshe and Aydin [q.vv.\ in 
1318-31 and 1345-60 resulted in limited damage (see 
E.A. Zachariadou, Trade and Crusade. Venetian Crete and 
the emirates of Menteshe and Aydin, 1300-1415, Venice 
1983, 13,51,93-4, 103, 188) and during the 9th/15th 
century, the Ottoman sultans recognised Venetian 
suzerainty over the island in a series of treaties (see 
Pitcher, Hist, geogr. of the Ottoman empire, 67 and maps 
VIII, XIV). Ottoman raids, including one by Khayr 
al-Din Pasha [q.v.] in 943/1537, led to an agreement 
( c ahd-ndma) in 972/1575, and in the next year Piyale 
Pash a [q.v.] took it in a series of operations which end- 
ed Latin rule in the Archipelago (see B. Slot, Ar- 
chipelagus turbatus ... 1550-1718, Leiden 1982, 32 ff., 
73 ft'.). In fact, it was administered on favourable 
terms by Joseph Nasi, Duke of Naxos [see nakshe] 
and in the following century was still largely self- 
governing but suffering from Venetian corsair raids, 
even though there were no Muslims on the island and 
at least part of the Christians were Roman Catholics. 
A brief Russian capture of the island during the 
1768-74 Russo-Turkish War was followed by 
definitive Ottoman rule until in May 1821 the island 
rebelled against Turkish rule (see Pegues, Histoire . . . 
de Santorin, Paris 1842, 352-3, 619 ff.), though, as was 
also the case in Naxos, by no means unanimously, 
given the Catholic population. The island in fact 
briefly rose against its first Greek governor, 
Capodistrias, in 1831, but was finally annexed to the 
newly-founded Kingdom of Greece by the 1832 Trea- 
ty of Constantinople. 

Bibliography (in addition to references in the ar- 
ticle): M. Danezes (ed.), Santorine, Athens 1971 (in 
Grk.); M. Medrinos, Turkish-dominated Thera. Its 
liberation movements (in Grk.), in EEKM, ix (1971-3), 
718-46; A. Tselikas, Memories from Santorine 1573- 
1819 (in Grk.), Athens 1985. (A. Savvides) 
SANUA L JAMES [see abu naddara]. 
al-SANUSI, Abu c Abd Allah Mahammad b. 
YQsuf b. c Umar b. Shu c ayb (b. at Tlemcen 838 or 
839/1435-6, d. Djumada II 895/May 1490), North 
African theologian and mystic. 
1. Life and influence. 

Despite the decline of the Banu Zayyan or c Abd al- 
Wadids [q.v.], Tlemcen was still one of the main 
cultural centres of the Maghrib. In his youth, al- 
SanusI studied the Kur'an with his father, and other 
teachers for Arabic language and for arithmetic and 
the law of successions (receiving an idjaza from one of 
the later teachers, Abu '1-Hasan al-Kalasadl, on 
whom see Brockelmann, IP, 343-4, SII, 378-9), for 
the Mudawwana of al-Tanukhl, for the astrolabe, for 
the usul al-din and for the Diurnal of al-Khunadji. In 
such an environment of scholars, al-Sanusi stood out 
as precocious. With his half-brother C A1T al-Talutl, he 
studied the Risala of Ibn Abl Zayd al-Kayrawanl 

[q.v.], the hshad of the Imam al-Haramayn al- 
Djuwaynl [q.v.] and tawhid, and received another 
idjaza. During a stay in Algiers, he studied the hadith 
collections under c Abd al-Rahman al-T_ha c alibi, and 
then at Oran followed the teaching of the great Sufi 
Ibrahim b. Muhammad al-TazT. 

Being inclined to asceticism, he fasted one day out 
of two and rarely left his dwelling. He kept vigils for 
several night on end and then embarked on a period 
of fasting. He was famed for his practise of istikhara 
[q.v.], one which enabled him to give a reply to a 
problem obtained during sleep. Because of his gift of 
knowing how to interpret (lit. "read", kara^a) his 
dreams and those of others, the public came to him 
from all over the kingdom to hear his dicta, and he 
gained an unequalled reputation in the religious, 
especially mystical, sciences. He continued to teach in 
the mosque until illness compelled him to slow down 
and in the end brought his life to an end in 894/1490. 

Soon afterwards, a kubba, rectangular in shape and 
covered with shiny green fabrics, was erected in his 
honour in the cemetery of al- c Ubbad (Sidi Bou- 
Medine); it contained the catafalques of al-Sanusi and 
his half-brother al-Taluti, who had died shortly after. 
But he allegedly had another tomb amongst the Awlad 
Hararau, thus gaining him the name of Bu Kabrln 
(likewise amongst the Ban! Bu Sa c Id, etc.). Each 
winter, when the corn was still green, a great com- 
munal meal (wa c da) was given by the people of 
Tlemcen to the poor and to strangers, near al-Sanusi's 
tomb, the latter being called Dhu 'l-Wakfa "master of 
the drought" because of a miracle described by Ibn 
Maryam in his Bustan. At the end of the meal, there 
was a communal prayer for rain. Sidi SanusI was 
equally invoked in periods of protracted drought. At 
Tlemcen, no less than two mosques perpetuate his 
memory. An apparently older one, at the gate of the 
Darb al-Masufa quarter, is said to be that of Ibn al- 
Banna 3 and to have been the mosque were al-Sanusi 
taught and worshipped. The other is said to have been 
built on the site of the house where he was born, in the 
Banu Djumla quarter. Both were richly endowed with 

Al-Sanusi had numerous students, such as al- 
Mallali, author of his biography and of a commentary 
on the $ughra; Abu '1-Kasim al-ZawawI; Ibn al- c Abbas 
al-Saghir, Ibn Abi Madyan; and Ibn Sa c d al- 
Tilimsanl, one of the sources of Ahmad Baba. They 
contributed to the spread of their master's works in 
the West Africa, by means of the trading links well es- 
tablished with that region, especially through the 
milieux of scholars, such as the family of Ahmad Baba 
of Timbuctu, in the 10th/16th century in Mali, and 
c Uthman b. FudI [q.v.] (Usumanu dan Fodio) in 
northern Nigeria in the 12th/18th century. 

Over the centuries, al-Sanusi has remained a source 
of inspiration and study. His three '■aka.Hd marked the 
grades of primary, middle and advanced studies, with 
the Sughra al-sughra sometimes replacing the Sughra in 
the first grade. In Algeria, there have been various 
commentaries from scholars so diverse as al- 
Warthilanl (d. 1193/1779), Abu Ras (d. 1238/1823), 
etc. The famous poet Ibn Amsayib (d. 1 190/1768) us- 
ed the theme of a visit to Sidi SanusI' s tomb and was 
buried near him. At the opening of the 20th century, 
Ibn Badls used al-Sanusi for his Kur'an commentary. 
Moroccan scholars studied the three '■aka}id with the 
commentaries and gloss of al-Dasukl. Thus the FasI 
al-Mandjur studied and commented on al-Sanusi, as 
well as the MiknasI al-Wallall, the minister al- 
Zayyanl and the Sufi Ibn c AdjIba. Till the beginning 
of the 20th century, the teaching of logic at Fas was 

done through the Mukhtasar; kaldm with the three 
creeds and the abridgement of the Sughrd, with com- 
mentaries; and theology with the Sughrd and the gloss 
of al-Yusi on the Rubra. In the 1 lth/ 1 7th century, the 
c Akida al-sughrd spread through West Africa to the 
Niger under the Fulani name kabbe. Thereafter, com- 
mentaries and glosses on them began to abound in the 
Arabic literature of West Africa, especially the lesser, 
middle and greater commentaries on the Sughrd, 
which was most often called Umm al-bardhin. 

In Egypt, al-Sanusi was taught at the Azhar, 
notably with the commentaries of al-Fadali (d. 1821) 
and his pupil al-Badjuri (d. 1860), up to the coming 
of the Salafiyya [q. v. ] cultural and social reform move- 
ment. Muhammad c Abduh [q.v.] often cites him, and 
it was certainly al-Sanusi who provided him with the 
three modalities of judgement: necessity (al-wudjub), 
impossibility (al-istihdla) and possibility (al-djawdz), 
and his attitude to the Muslim community, in his 
Risdlat al-Tawhid, has reminiscences of that of the 
scholar of Tlemcen. Finally, in Asia, and especially in 
Malaysia and Indonesia, it has been the Umm al- 
bardhin, also called al-Durra, which has been the most 
popular of those works explaining the Ashman doc- 
trine on the divine and prophetal attributes (sifdt). 
Copies of it often have interlinear Malay or Javanese 
translations. In the Pesantrens [q.v.], the commen- 
taries and glosses are studied on the old, original texts 

2. Works. 

As far as is at present known al-Sanusi's work 
(apart from the five '■akdHd, al-Rubrd, al-Wustd, al- 
Sughra, Sughrd al-sughrd and Mukaddima) comprises a 
large number of commentaries on many varied 
topics — works of logic, the Sahih of Muslim, algebra, 
medicine, the Kur'anic kira'at, Sufi manuals, to/sir, 
etc. (for details, see Brockelmann, II 2 , 323-6, S II, 
352-6, and Moh. Ben Cheneb, ET art. s.v.). Almost 
all of these are abridgements or works of vulgari- 

In his Sughrd, al-Sanusi gives an exposition of the 
essentials of faith according to a methodical form of 
augment. Knowledge ( c ilm) is based on judgement 
hukm), which stems from the divine law (sharfa), 
custom C-ada), experience and reason ( c akl). Rational 
judgement (al-hukm al- c akli) comprises the three 
categories of necessity, impossibility and possibility, 
Aristotelian modalities adopted by Islamic philosophy 
and theology. God has twenty attributes: six are 
divine, seven comprise real ideas (sifdt al-ma'-dni) and 
seven are ideal ones {sifdt ma c nawiyya). These are fol- 
lowed by the attributes of the divine messengers. 
Finally, he concludes that the shahdda sums up in its 
first part all that a believer need know of God, whilst 
the second part involves belief in all the prophets, 
angels, revealed books and the Resurrection, affirmed 
as true by the Prophet. 

Al-Sanusi's doctrine, as it appears from his works, 
is made up of prudence (tawakkuf) and reason, in pur- 
suit of equilibrium and salvation. It was certainly in- 
fluenced by signs of a shaking-up of values in the 
society of his time vis-a-vis the state, religion and 
cultural traditions, with the faith confronted by new 
problems. It was open to non-Muslim theological in- 
fluences, as in his interpretation of the Gospel accord- 
ing to St. John, xiv, 16-17, 26, 29, and of the theme 
of the Paraclete (baraklit): the saviour of the human 
race is our Lord Muhammad (rahma li 'l- c dlamm). The 
ideas of the various madhdhib are set forth and dis- 
cussed. It was also inspired by the works of al- 
GhazalT's teacher al-Djuwayni (the former's ideas 
being spread in the Maghrib by Ibn Tumart [q.v.]), 

notably by the Irshdd and the Shdmil, seen both in its 
form and in certain concepts, such as al-Ash c arI's 
classification of the divine attributes into three sub- 
divisions, as in al-Djuwayni; and the difference be- 
tween nafsiyya or consubstantial attributes and 
qualitative, ma'nawiyya, ones. Similarly, al-Ghazali's 
influence (the Iktiidd) appears in the division into 
seven of the divine attributes of God and the prophets, 
whilst al-Sanusi also takes up the Ash c ari idea of kasb. 

However, at times he takes up a position differing 
from such Ash c arl authorities as these two, al-Ash c ari 
himself and al-Bakillani, rejecting that which offends 
against reason and dogma (Mukaddima, 65-7). His 
system of ethics is inspired by al-Ghazall. especially in 
the Ihyd?, where, on all occasions, he seeks out a mid- 
dle position. 

He condemns firmly blind acceptance (laklid [q.v.]). 
Merely reading the Kur'an and hadiths is inadequate 
for understanding the bases of faith (imdn); sound 
reasoning (al-nazar al-sahih) is preferable to this 
(Wusta), and must lead to what is sought. Judgement 
thus rests on proof, aided by knowledge (Him) or intui- 
tion (ma'-rifa). Dogmatic theology is the queen of 
sciences and involves demonstrable proof (burhan) to 
distinguish between dialectic, rhetoric, poetry and 
sophism (the divisions of Aristotelian logic). The 
truths of the Kurgan can thereby be apprehended and 
anthropomorphic interpretations (tadjsim) avoided. 
Following a tradition of the Almohad dynasty, al- 
Sanusi wanted the people to have a simplified access 
to God, leaving for an elite (khdssa) the possibility of 
a deep study of the principles of religion, a theme 
which he often takes up with his pupils and in his 
works, notably in his division of knowledge into two 
branches, an external (zdhir) one and an esoteric 
(bdtin) one, for him the truest and noblest knowledge. 

Finally, by his public and private life, al-Sanusi 
became part of the popular Sufism illustrated by 
Tlemcen's patron saint, Abu Madyan: a moderate 
Sufism, with simple dogma and intelligible to the 
believers. From his independence and originality, and 
his deep, widely-connected thought, he seems clearly 
to be, in company with the Persian al-Dawam [q.v.], 
the most important Muslim theologian of the 9th/ 15th 
century (Fakhry). 


. The 

Ibn c Askar, Dawhat al-ndshir, ms. B.N. 5025, Fas, 
1309, 89, tr. Graulle, in AM, xix (1913), 207-12, 
221, 29; Muhammad al-Mallall, al-Mawdhib al- 
kudsiyyafi 'l-mandkib al-sandsiyya, ms. B.N. 6897, 
B.N. Algiers 1066, etc. Of secondary importance 
are c Abd al-BasIt b. Khaffl, al-Rawdal-basim, ed. and 
tr. R. Brunschvig, Deux recils de voyage inedits en Afri- 
que du Nord au XV Steele, ^Abdalbasit et Adorne, Paris 
1936; Ahmad Baba al-Tinbuktl, Nayl al-ibtihadj, Fas 
1309; idem, Rifdyat al-muhtddj; Ibn Maryam, al- 
Bustdn fi dhikr al-awliyd' wa 'l-^ulamd' bi-Tilimsdn, 
Algiers 1908, 198, tr. F. Provenzali, Algiers 1910, 
270-84; WansharishI, tr. E. Amar, in AM, xii 
(1-908), 82-3, 265; al-Hasan al-Wazzan (Leo 
Africanus), Description de I'Afrique, tr. Epaulard, 
Paris 1952. 

2. Studies. J.J. L. Barges, Hist, des Beni-Zeiyan, 
rots de Tlemcen, Paris 1852; idem, Tlemcen, ancienne 
capitate du royaume de ce nom, Paris 1859; idem, Com- 
plement de I'hist. des Beni-Zeiyan, Paris 1887; Moh. 
Ben Cheneb, Etude sur les personages mentionnes dans 
I'idjaza du cheikh c Abd el Qadir al-Fasy, in Actes XIV' 
Congres internat. des orientalistes , Algiers 1905-Paris 
1907, iv; C. Brosselard, Les inscriptions arabes de 
Tlemcen. Tombes du Cid Mohammed es-Senouci et de son 
jure ..., in RAfr, iii/16 (1859), 245-8; idem, Retour 

a Sidi-Senouci, in ibid., v/28 (1861), 21-60; idem, In- 
scriptions — habous des mosquees de Sidi-Senouci, ir 
ibid., v/29 (1861), 321-36; Djamal al-DIn Bukli, al 
Imam Ibn Yusuf al-Sanusi wa- c ilm al-tawhid, Algier: 
1985; A. Cherbonneau, Docs, inedils sur es-Senouci, ir 
JA, iii (1854), 175-80; idem, Les krivains de I'Algeru 
auMoyen-Age, in RAfr, xiv (1870), 72-8; G. Delphin 
La philosophic du cheikh Senoussi d'apres son '■Aqida es 
su'ra, in JA, x (1897), 365-70; G. Gabrieli, Ut 
capitulo di teodicea musulmana, ovveri gli attribuli divin. 
secundo la Umm al-Barahin di al-Sanusi, Trani 1901 



n Bessc 

ii (1904); L. Gardet, Raison et Joi dans I'Islam, in 
Rev. Thomiste, xliii/3 (1937); idem, Etudes de 
philosophic et de mystique comparees, Paris 1972; D. 
Gimaret, Theorie de facte humain en the'ologie 
musulmane, Paris 1980; Abu '1-Kasim Muh. al- 
Hifnawl, Ta<rif al-khalaf bi-ridjal al-salaf, Algiers 
1324/1906; M. Horten, Sanusi und die griechische 
Philosophic, in Isl., vi (1915), 178-88; idem, Muham- 
medanische Glaubenslehre. Die Katechismen des Fudali und 
des Sanusi, Bonn 1916; J. P. Kenny, Muslim theology 
as presented by ... as-Sanusi, especially in his al-Aqida al- 
Wusta, Ph.D diss. Edinburgh 1970; J.D. Luciani, 
Petit traite de theologie musulmane, par ... Senoussi, 
Algiers 1896; idem, A propos de la traduction de la 
Senoussia, in Rafr, xlii/231 (1898), 376-88; idem, Les 
prolegomenes theologiques de Senoussi, Algiers 1908; 
T.H. Weir, The Shaikhs of Morocco in the XVIth cen- 
tury, Edinburgh-London 1904; M. Wolff, El- 
Senusi's Begriffsentwicklung des Muhammedanischen 
Glaubensbekenntnisses , Leipzig 1848; M.A. Zouber, 
Ahmad Bdbdde Tombouctou (1556-1627), Paris 1977. 
On the influence and diffusion of al- 
Sanusi 's teaching through the Islamic world, 
see: O. Amin, Muhammad '■Abduh, essai sur ses idles 
..., Cairo 1944; A. Bel, in Bull, de I'enseignement des 
indigenes de I'Academie d'Alger, clxxxiv (1908); M. 
Ben Cheneb, Itineraire de Tlemcen a la Mekke, in RAfr, 
xliv (1900); L.W.C. van den Berg, in BTLV 
(1886), 538-42; J. Berque, Al-Yousi ..., Paris-The 
Hague 1958; M. van Bruinessen, in BTLV, cxlvi/2- 
3 (1990), 251-2, 265; A. Cabaton, Une traduction in- 
terlineaire malaise de la ^Aqidah d' al-Sanusi, in J A 
(1904), 115-45; M.A. Cherbonneau, Hist, de la lilt, 
arabe au Soudan, inJA, vi (1855); idem, Essai sur la 
litl. arabe au Soudan, in Annuaire de la Soc. Archeol. de 
... Constantine (1854-5); G. Delphin, in Bull. 
Trimestr. de Ge'ogr. et d 'Archeol. d'Oran, viii (1888); G. 
Faure-Biguet, iaJA, xiv, 2 (1899); M. Fakhry, A 
history of Islamic philosophy, London 1983, 322, Fr. 
tr. Histoirede la philosophic islamique, Paris 1989, 350; 
J.O. Hunwick, in BSOAS, xxvii (1964); W.E.N. 
Kensdale, Field notes on the Arabic lit. of the Western 
Sudan: Shehu Usumanu dan Fodio, inJfi/lS'(1955); P. 
Marty, Etudes sur I'Islam au Senegal, Paris 1917; 
idem, Etudes sur I'Islam et les tribus du Soudan, Paris 
1920; idem, Etudes sur I'Islam en Cote d'lvoire, Paris 
1922; A. Merad, Ibn Badis, commentateur du Coran, 
Paris 1971; I. Salam, L'enseignement islamique en 
Egypte, Cairo 1938; J.S. Trimingham, Islam in the 
Sudan, Oxford 1949; Warthilam, Rihla, partial tr. in 
M. Hadj-Sadok, in RAfr, xcv; Ziyanl, al- 
Turdjumdna al-kubrd, partial tr. G. Salmon, in AM, 
ii (1905). 





Bel, Quelques rites pour obtenir la pluie ..., in Recueils de 
memoires ... XIV' Congres des Orientalistes , Algiers 
1905; idem, La population musulmane de Tlemcen, in 
Rev. des Et. Ethnographiques et Sociologiques -(1908); 
idem, Les Beni Snous et leurs mosquees, in Bull. Archeol. 
(1920); idem, L'Islam mystique, Algiers 1928; G. 

Destaing, Etude sur le dialecle Berbere des Beni-Snous, in 
Bui. de Correspondance Africaine, xxxiv (1907); G. 
Marcais,_ Tlemcen, Paris 1950. (H. Bencheneb) 
al-SANUSI, Muhammad b. c Ali, scholar and 
Sufi of Algerian origin, and founder of the 
■Sanusiyya [q.v.]. 

Born on 12 Rabi< I 1202/22 December 1787 at al- 
/Vasita (or douar al-Thorch), some 40 km/25 miles to 
he east of Mustaghanim, he belonged to the Awlad 
Sldi c Abd Allah, Idrisid shurafP of the Madjahir, a 
Makhzan tribe of Oran under the Turkish regime. He 
inherited from his family the nisba al-Khattabl (from 
<Abd Allah al-Khattab, the ancestor of the Awlad Sldi 
c Abd Allah, who lived at the beginning of the 
10th/16th century), and those of al-Hasanl and al- 
Idrisi (which refer to the origins of his Sharifian line). 
The qualificative of al-SanusT would appear to have 
been borrowed from a local toponym (the name of a 
jntain) of the region of Tilimsan (Tlemcen), 
:re his family previously lived. This nisba implies 
link of kinship with Muhammad b. Yusuf al- 
Sanusi, the Tlemceni scholar of the 9th/14th century 

Muhammad al-Sanusi studied initially within his 
own family and in several intellectual centres of Oran: 
Mustaghanim, Mazuna and Tilimsan. At eighteen 
rs old, in 1220/1805 (al-Dadjdjani, following King 
Idris), or at twenty-one, in 1223/1808-1809 (K. 
Vikor, following al-LIbi), he left to pursue his studies 
t Fez. He stayed there, according to the authors, 
:ven years (al-Libi/Vikor) or fourteen years (Idris/al- 
Dadjdjani), associating with major scholars of the 
period. In Fez, he joined several major Sufi orders, in 
particular the Shadhiliyya in its Djazuliyya, Nasiriyya 
and Tayyibiyya branches, as well as various local 
turuk. He appears also to have been affiliated to the 
Darkawiyya and the Kadiriyya, but he does not men- 
tion them in the list of the forty turuk, the wird of which 
he was later to claim. He was the pupil of Ahmad al- 
Tidjam, with whom he studied the Kur'an, but whose 
way he does not seem to have adopted. 

He left Fez in 1233/1818 (al-LIbi) or in 1235/1819- 
20 (al-Dadjdjani). According to al-LIbi, he apparently 
made a first pilgrimage to Mecca in 1232-3/1817-18, 
before his final departure from the Moroccan city. 
After Fez, his traces are followed with some difficulty: 
he probably spent some time in his native region, and 
then he made a leisurely progress along the confines 
of the Sahara. But the statement of Rinn (1884), ac- 
cording to which al-Sanusi was crossing the Sahara at 
the time of the arrival of the French in Algiers, is not 
to be taken seriously. He had long since left Algeria 
and was never to return there. 

After staying two to three years in Cairo, al-Sanusi 
arrived in Mecca ca. 1240-1/1825-6. It was there that 
he encountered the Moroccan scholar and Sufi 
Ahmad b. Idris, who had been residing in the Holy 
City some twenty years. A mystical master, Ibn Idris 
claimed a brief lineage which was traced by way of 
two intermediaries to the mysterious Kur'anic figure 
of al-Khadir [q.v.], and hence to the Prophet, always 
present at the origin of these successive transmissions 
(whence the names of Khadiriyya and Muhammadiya 
given to the Way). Ibn Idris was not the founder of a 
structured fraternity but the respected master of a cir- 
cle of disciples in the Holy Cities, and later in the 

Al-Sanusi did not remain for more than two or 
three years under the tutelage of his master. When the 
latter departed for the Yemen (1243/1827-8), he 
became his khalifa at Mecca. As a pupil of Ibn Idris, 
al-Sanusi had acquired specific esoteric knowledge 

and had reconsidered certain areas of Him. It was also 
no doubt under his influence that he consolidated his 
distinctive positions in judicial matters: the criticism 
of blind imitation (taklid), the combined utilisation of 
the four madhahib, and the recourse to idjtihad, which 
were to characterise profoundly his later positions, 
and which would expose him to the criticisms of the 
scholars of al-Azhar (in particular, of Muhammad 
c Ali c IllaysJi). The teachings of Ibn Idris were clearly 
distinct from those of the Wahhabis, with whom a 
number of restrained polemics were held. 

When Ibn Idris died at Sabya in 1253/1837, al- 
Sanusi became a master in his own right and the 
zawiya of Ibn Idris, situated on the hill of Abu 
Kubays, a neighbourhood of Mecca, became by 
gradual and barely perceptible stages, at no date 
which can be specifically assigned, the centre of the 
new "Sanusiyya" brotherhood. Distinct derivatives 
emerged at the same time under the influence of other 
disciples, such as the Mirghaniyya which found its 
preferred territory in the Sudan. 

Probably hindered in Arabia, by Wahhabi activity 
and the power of the local religious establishment, the 
founder of the Sanusiyya henceforward sought new 
places for settlement in the African continent. From 
this time until his death, the chronology of his life is 
one of alternation between Mecca and the new 
African zawiyas: 1255/1840: departure from Mecca; 
1262/1846: return to Mecca, after founding four 
zawiyas, including that of al-Bayda 3 , in the territory of 
Barka (Cyrenaica); 1270/1854: new departure from 
Mecca and settlement at al- c Izziyya, then, in 
1273/1856, at Djaghbub, a "new city" created out of 
nothing in an inhospitable desert location on the 
western borders of Egypt. It was there that he died, on 
9 Safar 1276/2 September 1859. 

At the time of his first journey to the West, 
Muhammad al-Sanusi had entered Tunisia and ap- 
proached the Algerian frontier, but the region was 
unstable at that time and seemed to him unfavourable 
for his projects, and the presence of other Sufi orders 
was also a hindrance. Finally, in Tripoli, he reached 
an agreement with the local Ottoman power, then 
barely reinstated in Libya, and he turned, in the 
course of his wanderings, towards the Djabal al- 
Akhdar, to the north of Cyrenaica, approaching the 
Bedouin of the region, entirely ignorant of the 
precepts of the Law. Thus the Sanusiyya, unable to 
establish itself in the neighbouring countries, became 
"Libyan", with a strong concentration in the Djabal 
al-Akhdar. and a salient extending across the Fezzan 
as far as the Algerian borders. 

Muhammad al-Sanusi conceived his work as being 
primarily a carefully-structured missionary project. 
He thus laid the foundations of a system which his son 
and successor Muhammad al-Mahdi was subsequent- 
ly to pursue energetically. In this system, the epitome 
of pious endeavour is the construction of a zawiya, a 
place of assembly, of teaching and of arbitration, a 
sedentary and agricultural centre for the diffusion of 
Sanusi ideas and policies in a nomadic environment. 
Thus there was established across Cyrenaica and the 
Fezzan a network of Sanusi zawiyas, which guaranteed 
the diffusion of Islamic models and celebrated the 
baraka of the founder. 

On the doctrinal level, al-Sanusi is seen as a refor- 
mist (appeal to idjtihad). Tenacious legends notwith- 
standing, al-Sanusi was not, no more than was 
Ahmad b. Idris, an apologist of djihad against the 
Europeans. He maintained an apolitical stance, 
prepared to deal with the Ottoman authorities and 
giving priority to a "grass-roots" re-Islamisation 

among the disinherited populations. In terms of 
Sufism, he is seen as the reviser, and to a certain ex- 
tent the unifier, of the forty turuk, the asanid of which 
he lists meticulously. Al-Sanusi left behind him, in the 
various domains of Him, and more particularly in that 
of Sufism, a prolific written corpus, quite apart from 
his numerous letters. Al-Dadjdjanl has listed 44 titles. 
Among these, eight of the most important have been 
collected and printed in al-Madjmu'at al-mukhtara at the 
orders of King Idris: al^Durar al-saniyya, which 
eulogises the IdrisI family; lkaz al-wasnan and Bughyat 
al-makasid, which deal with taklid and idjtihad; K. al- 
Manhal al-rawi al-ra'ik, which, in its early sections, 
recalls his masters and his studies; Mukaddimat 
Muwatta - 1 al-lmam Malik, which addresses questions of 
Malikllaw; al-Salsabil al-muHn, which is devoted to the 
analysis of forty Sufi brotherhoods; and ShifP al-sadr 
bi-ara al-masaHl al-'ashar, which concerns ten problems 
of cultic practice (prayer with arms folded, etc.). 
Some works, destroyed by the Italians, are known on- 
ly by their titles. 

Thus there evolves, across the 72 years of his life, 
the figure of a teacher versed in multiple disciplines, 
of a reviver of Islam and of Sufism, of a missionary 
and an organiser, but also of an eminent individual, 
endowed with baraka, whose charisma was renowned 
throughout the central and eastern Sahara. 

Bibliography: Muhammad al-Tayyib b. Idris al- 
Ashhab, al-Sanusi al-kabir , Cairo 1956; Ahmad Sidkl 
al-Dadjdjanl, al-Harakat al- sanusiyya. NasPatuhd wa- 
numuwwuha fi 'l-karn al-tasi c <ashar, Beirut 1967, 
2 1988; E.E. Evans-Pritchard, The Sanusi of Cyre- 
naica, Oxford 1949, 2 1954; c Abd al-Malik b. c Abd 
al-Kadir b. c Ali al-LIbl, al-Fawa^id al-djaliyya ft 
tahikh al-Vila al-Sanusiyya al-hakima fi Libiyya, 
Damascus 1966; B.G. Martin, Muslim brotherhoods 
in 19th-century Africa, Cambridge 1976; R.S. 
O'Fahey, Enigmatic saint. Ahmad Ibn Idris and the 
Idrisi tradition, London 1990; Muhammad b. c AlIal- 
SanusI, al-Madjmu c at al-mukhtara min mu^allafdt al- 
ustadh... al-SanuSi, Beirut 1968; L. Rinn, Marabouts 
et Khouan, Algiers 1884; C. Souriau, [Muhammad b.] 
<Ali al-Sanusi (1787-1859), fondateur d'une confrerie aux 
dimensions d'un empire, in Ch.-A. Julien (ed.), Les 
Africains, Paris, vi (1977), 231-59; J.-L. Triaud, La 
legende noire de la Sanusiyya, Paris and Aix-en- 
Provence 1995; K.S. Vikor, Sufi and scholar on the 
desert edge. Muhammad b. '■All al-Sanusi and his 
brotherhood, London 1995. (J.-L. Triaud) 

al-SANUSI, Shaykh Sayyid Ahmad (1290-1351/ 
1873-1933), Third Grand Master of the Sanu- 
siyya [q.v.] order of dervishes. His full name was al- 
Sayyid Muhammad al-Mahdi Ahmad b. al-Sayyid 
Muhammad al-Sharlf b. al-Sayyid Muhammad b. 
C A1I al-Sanusi al-Khattabi al-Hasani al-Idrisi. 

He was born at Djaghbub, a grandson of the 
founder of the order, and received a classical educa- 
tion in Islamic learning from his father, uncle, etc., 
according to the high standards of Sanusi tradition. 
He succeeded to the leadership at Kuru (Borku, in 
Chad), where his uncle resided from 1899 till his 
death in 1902, being the eldest member of the Sanusi 
family. Next to his spiritual leadership, Ahmad al- 
Sharif developed a political and military organisation 
for the Sanusi community against French expansion 
in the Sudan region, but after a defeat, he decided to 
withdraw from Kuru to the old centre at Kufra [q. v. ] 
in 1902. In need of international recognition and sup- 
port, he agreed to the establishment of direct Ottoman 
rule in Cyrenaica and Fazzan (1910), and the 7,000 
Ottoman troops in the province co-operated now 
against the enemies of the faith, these being in 1911 


the Italians invading Ottoman Libya [see libiya. 3] 
until Muharram 1331 /December 1912. During this 
period the Sanusi shaykh issued a proclamation of 
djihad against the enemies of Islam. After the conclu- 
sion of the Italo-Ottoman peace of Lausanne-Ouchy 
(15 October 1912), the Sanusi leader continued 
resistance against the Italians with the discreet sup- 
port of the Ottoman (Young Turks') government. 
The sultan-caliph, Mehemmed V [q.v.] approved of 
Ahmad al- Sharif s installation of a "Sanusiyya 
Government" in Cyrenaica and Fazzan. From 1912 
till 1915 Ahmad al-Sharif was able to defeat Italian 
forces at times. Apart from regular financial and 
logistic support, the sultan-caliph awarded the Shaykh 
honours and decorations. When Italy joined the 
Entente Powers against, i.a., the Ottoman Empire in 
1915, Ahmad al-Sharif was secretly appointed the 
sultan's representative (nd'ib ul-sultan) with the rank of 
Vizier and the title of Pasha (irade of 6 August 1915). 
During the same year 1915, from June onwards, a 
regular communication with Istanbul was ensured by 
German and Austro-Hungarian submarines carrying 
arms, munitions and men to the Sanusi forces, but 
their guerrilla attack against the British in western 
Egypt failed (15 Rabl* I 1334/22-3 January 1916). 

Ahmad al-Sharif retreated with 800 followers and 
was chased back to Djaghbub, from where he went on 
to the Sirtica region of Tripolitania. He maintained 
his relationship with the Ottomans, but the influence 
of his cousin, Sayyid Muhammad Idris (the later King 
Idris I al-Sanusi of Libya) was by now increasing. 
With the permission of Ahmad al-Sharif, Idris opened 
up negotiations with the British and the Italians 
(1917). In 1918 the Sanusi shaykh was made the 
sultan's representative in all North Africa, but his ac- 
tual influence on affairs was steadily in decline. On 13 
Dhu '1-Ka c da 1336/21 August 1918 Ahmad al-Sharif 
left Libya, for good as it turned out to be, brought by 
a German submarine to Istanbul. He relinquished 
political leadership but remained the spiritual chief of 
the Sanusiyya tarika till his death, and his lasting 
prestige is evident from the fact that he was chosen to 
officiate at the ceremonial girding of the sword of the 
new Ottoman sultan, Mehemmed VI Wahid al-Dln 
[q.v.], at Eyyub in 1336/1918. Widely regarded as one 
of the foremost fighters for Islam, Ahmad al-Sharif 
chose the side of the resistance against the Allies in 
Anatolia led by Mustafa Kemal Pasha [see ataturk], 
and became one of his emissaries in the provinces of 
Anatolia. (Mustafa Kemal was photographed in an 
Arab dress presented to him by the Sanusi Shaykh at 
this time.) In 1922 he journeyed to southeastern 
Anatolia along the Turco-French front, inter alia ar- 
bitrating a peace amongst the Arab tribes there, but 
after the definitive victory of Mustafa Kemal, Ahmad 
al-Sharif returned to Istanbul and became involved in 
the question of the Ottoman caliphate [see khilafa], 
which was not after all offered to him, in spite of the 
support of his cause by Ibn Su c ud [see c abd al- c aziz 
al sa'ud], Imam Yahya of Yemen and Sa c d Zaghlul 
Pasha [q.v.]. In 1924 he left for Damascus. The 
French did not permit him to stay there, and he went 
on to the Hidjaz, dying at Medina on 13 Dhu '1- 
Ka c da 1351/10 March 1933. 

Bibliography: Ahmad al-Sharif. Bughyat al- 
musaHdfi ahkdm al-mudjahid fi 'l-hathth c ala 'l-djihad, 
Cairo 1332/1913-14; S.S. Aydemir, Enver Pasa II 
1908-1914, Istanbul 1971, 229-40; E.E. Evans- 
Pritchard, The Sanusi of Cyrenaica, Oxford 1949 (por- 
trait between 128-9); M. Le Gall, The Ottoman 
government and the Sanusiyya: a reappraisal, in IJMES, 
xxi (1989), 91-106; E. Graefe, Der Aufruf des Scheichs 

der Senusija zum Heiligen Knege, in Isl., iii (1912), 
141-50; Jaeschke and E. Pritsch, Die Tiirkei seit dem 
Weltkriege. Geschichtskalender 1918-1928, in WI, x 
(1927-9), 1-154; E. Kedourie, Egypt and the caliphate, 
1915-52, in The Chatham House version and other Mid- 
dle Eastern studies, London 1970, 189, 423; O. 
Kologlu, Mustafa Kemal'in yanmda iki Libyah lider, 
Ahmet Serif-Suleyman Baruni, Ankara 1981; J.M. 
Landau, The politics of Pan-Islam, Oxford 1990, 134- 
8, 233; R. Peters, Islam and colonialism, The Hague 
1979, 86-9; R. Simon, Libya between Ottomanism and 
nationalism: the Ottoman involvement in Libya during the 
war with Italy (1911-1919), Berlin 1987, index s.v. 
Ahmad al-Sharif; S.R. Sonyel, Turk kurtulus savasi 
ve dis politika I, Ankara 1973, 188-9; N.A. Ziadeh, 
Sanusiyah, Leiden 1958, index s.v. Sayyid Ahmad 
ash-Sharif. (A.H. de Groot) 

SANUSIYYA, a Sufi brotherhood or tarika, 
founded in the middle of the 19th century by Muham- 
mad b. C A1I al-Sanusi [q.v.]. Denounced by a French 
"black legend", as tenacious as it is ill-founded, as a 
centre of anti-western subversion across the Sahara, 
the Sanusiyya order was confronted at a very early 
stage with the game of the European Powers. For this 

supposed, than for its specifically Sufi teaching. 

After studying in Fez, Muhammad b. C A1I al- 
SanusI (born in December 1787 near Mustaghanim, 
in Algeria), had become, at the time of his Pilgrimage 
during the 1820s, one of the disciples of Ahmad b. 
Idris, a Moroccan Sufi and Sharif residing in the Holy 
Places; when the latter departed for the Yemen in 
1827, he remained in Mecca as his khalifa. Ahmad b. 
Idris belonged to a mystic lineage going back to the 
Moroccan Sufi and Sharif c Abd al- c Az!z al-Dabbagh, 
belonging to a Shadhill spiritual lineage, initiated by 
the mysterious Kur'anic figure known by the name of 
al-Khadir [q.v.]. 

After the death in 1837 of Ahmad b. Idris, who was 
more the founder of a circle of disciples than of an 
organised tarika, the disciples of the Master split into 
different groups. Muhammad b. C A1I al-Sanusi, who 
was one of the oldest and, perhaps, one of the most 
advanced in the way, then inaugurated his own struc- 
ture by gradual stages, and the Idrlsiyya zdwiya of 
Mecca, on the hill of Abu Kubays, became the first 
"Sanusi" establishment. 

Muhammad al-Sanusi emerges, in his works, as the 
inspirer of a SunnI and moderate Sufism. Summing 
up the forty ways into which he was initiated, he 
claims a tarika muhammadiyya which is the quintessence 
of them all and which makes the encounter, in dreams 
and while awake, with the Prophet, and the appeal for 
his guidance and imitation of his model, the supreme 
qualities of the initiate. The dhikr of the (artka is con- 
stituted by a distinctive prayer, the saldt al- c azimiyya, 
a prayer for the Prophet inherited from Ahmad b. 
Idris, which takes its title from the repetition of Allah 
al-'-Azim ("God the Most Great"), and by various 
ahzdb and awrdd, including a haylala the text of which 
is prolonged by the formula "in every look and every 
breath, a number of times which only the knowledge 
of God can apprehend" (a formula which also features 
in the '■AzTmiyya). 

But it is not the dhikr which constitutes the most 
original part of the Sanusi programme. The principal 
work of Muhammad al-Sanusi was directed in fact 
towards the realisation of a centralised and hierar- 
chical order based on a network of zdwiyas established 
in places judged to be strategic from the point of view 
ication routes, of the supply of water or of 
ibal composition. The zdwiya is thus the 


ultimate act of piety, and the creation of it follows a 
precise protocol. The appeal for it must be made by 
the population concerned. The latter sends a delega- 
tion to the chief of the brotherhood, before whom it 
solemnly confirms its wishes, and takes on the obliga- 
tion of constructing the buildings of the zawiya and of 
working periodically in its service. Muhammad al- 
Sanusi would then appoint, from among his disciples, 
a shaykh, obliged to marry into the local population. 
Regular and prolific correspondence maintains the 
link between the centre and the new local outpost. 
Each zawiya is simultaneously an educational centre, 
a staging-post and a hostel for travellers, a place of 
worship and an agricultural location. It is, in the 
desert environment, a small urban enclave, represen- 
ting the sedentary values and models which are those 
of normative Islam. By virtue of this network, which 
developed in Libya from the 1840s onward, the tarika 
is observed, in the first phase of its existence, to be a 
missionary order whose vocation is then to 
disseminate Islam among the disinherited populations 
of the central Sahara. 

The history of the SanQsiyya may be divided into 
several sequences. The first, under the leadership of 
the founder — known as al-Ustadh (Master) and as al- 
Jmdm al-akbar (Great Imam) — saw the expansion of 
the movement in an east-west direction, following the 
lines of wells and the routes of the Pilgrimage. Before 
reaching its furthest point and establishing its niche in 
Cyrenaica, the brotherhood had, in fact, spent a long 
time in search of the right territory. The SanQsiyya 
often encountered the resistance of the existing local 
powers and the presence of other Islamic organisa- 
tions and clienteles. Its extension into Cyrenaica 
(foundation of the first African zawiya, al-Bayda 3 , at 
Cyrene at the end of 1842) and, to a lesser extent, into 
the central Sahara, thus represents a last resource: it 
was only in these regions that the presence of weak 
and scattered authorities, and the absence of powerful 
religious institutions, enabled it to acquire suitable 

During its expansion in the Sahara, the SanQsiyya 
came into contact with the world of the nomad. It was 
there, far from the major centres of power and of 
scholarship, that it found a loyal following, one which 
was furthermore more attached to the baraka of the 
Master than to his erudite teaching. Thus there was 
established between the brotherhood and the Saharan 
nomads a durable bond of friendship. The Sanusi 
made alliance with the Bedouin (Madjabra and 
Zwaya, in Cyrenaica, and to a lesser extent, Twareg 
Ajjer, to the west) one of the pillars of their system. 

But the Sanusi system was not restricted to the 
nomadic world. The brotherhood exploited to its ad- 
vantage a new trans-Saharan route, inaugurated at 
the beginning of the century through the initiative of 
Ouaddai (Waddai). This route, still precarious and 
experimental, which ran from Benghazi to Abeche by 
way of Kufra, and which was one of the last trans- 
Saharan axes still usable (the others having fallen vic- 
tim to political anarchy or to European interference), 
became under its protection, from the middle of the 
century onward (zawiya of Tazar, to the north of 
Kufra, 1848-9; zawiya of al-Djuf, Kufra, ca. 1856), 
one of the principal foundations of its power. The 
Sanusi system was born of this combination of ex- 
ploitation of a caravan axis with colonisation of the 
desert. In the course of this process, the tarika became 
the manager and controller of a region and the pro- 
moter of an economy. But its activity was essentially 
regulatory. Just as it did not then aspire (o political 
power, the SanQsiyya had no wish to undertake 

economic enterprises. The benefits which accrued to 
it from its protection of commerce were symbolic 
(alliances and allegiances) or material (gifts, agri- 
cultural produce). The activity of the Sanusi ikhwan 
was one element in a major project, envisaging first a 
general recognition of the baraka of the Master, from 
which esteem, clientele and material goods would 
subsequently flow. 

A veritable "Maghrib! multi-national" at the 
outset, bearing in mind the origin of the closest 
disciples of the Master, the SanQsiyya then began to 
"Libyanise" itself. The two sons of the Imam al- 
akbar' s middle age were born in Cyrenaica, respective- 
ly in 1262/1844 (Muhammad al-Mahdl) and 
1262/1846 (Muhammad al-Sharif). Muhammad b. 
C A1I al-SanusI himself left the Hidjaz for good in 1854 
and established his headquarters at Djaghbub, a new 
city in the heart of the desert and near the Egyptian 
border, in 1856. It was there that he died in 1859. 

His son and successor Muhammad al-Mahdl, who 
was to preside over the destinies of the movement for 
more than forty years, emerges as an organiser of 
talent. It was he who gave to the foundations laid by 
his father a systematic development. It was also he 
who oriented the brotherhood towards the south, in 
the direction of Central Africa. The first sub-Saharan 
zawiya came into being at Chemidour (currently in 
eastern Niger) from 1861-2 onwards. Later, faced by 
increasing interference on the part of c Abd al-Hamid 
II, Muhammad al-Mahdl who, like his father, had 
maintained amicable relations with the Ottoman 
authorities, abruptly transferred his headquarters 
from Djaghbub to Kufra (June 1895), then to Gouro, 
to the north of what is currently Chad, in December 
1899, preferring a hidjra to confrontation. The attrac- 
tions of Ouaddai, rich in ivory, in ostrich feathers 
and, additionally, in slaves, also played a role in this 
long march towards the south. 

This southward orientation of the movement coin- 
cided with the French advance towards Lake Chad. 
From 1901 onwards the SanQsiyya improvised, in dif- 
ficult conditions, resistance to the French assaults, 
establishing for this purpose a defensive system and 
then appealing for Turkish protection. Its destiny was 
henceforward to be inseparably embroiled in the game 
of the Great Powers. Thus on 9 November 1901, 
French troops attacked the zawiya of BIr c Alali situated 
some 100 km/60 miles from Lake Chad. Initially 
defeated, the French forces took control of the place 
on 20 January 1902. This was the beginning of a long 
Franco-Sanusi war which ended with the fall of the 
fortified zawiya of c Ayn Galakka (to the south of 
Gouro) on 27 November 1913. 

After the death of Muhammad al-Mahdl at Gouro 
on 2 June 1902, the latter's nephew, Ahmad al-Sharif 
(1872-1933 [see al-sanusI, shaykh sayyid ahmad]), 
became the third Master of the tarika. He immediately 
decided to return to Kufra and began to organise the 
brotherhood on more secular lines, a development 
which prolonged and reinforced the politicisation and 
militarisation which had become apparent during the 
confrontation with the French. When the Italians ar- 
rived in Libya (October 1911), he used all his influ- 
ence to achieve the mobilisation of his followers in a 
djihad against the invader and allied himself with the 
Ottoman forces. This "djihadist" orientation, alien to 
the founder of the Way, henceforward made of the 
SanQsiyya a politico-military organisation aligned 
with the Ottoman caliphate. In 1914, Ahmad al- 
Sharlf allied himself to the Central Powers. At the lat- 
ter's insistence, Sanusi forces fought on all fronts: 
they attacked the British in Egypt (November 1915), 


expelled the Italians from Tripolitania and the Fezzan 
(September 1914-April 1915), then engaged in a con- 
flict with the French in the Sahara, which culminated 
in the seizure of the French fortress of Djanet (2 
March 1916), the capture of P. de Foucauld, killed ac- 
cidentally by a Twareg sentry (1 December 1916), 
and the siege of the French outpost of Agades (1 
December 1916-3 March 1917). Already split into a 
number of "fiefs" according to the various branches 
of the family, deprived progressively of a single direc- 
tion, the Sanusiyya were devastated by this war and 
by the defeats inflicted on it in all theatres of operation 
after 1916. Muhammad Idris, elder son of Muham- 
mad al-Mahdl, promoted by the British as a useful in- 
termediary, negotiated with the latter and with the 
Italians and signed an accord at c Akrama near 
Tobruk in April 1917. The accord integrated the 
Sanusis into the camp of the Allied Powers in ex- 
change for a partial recognition of the brotherhood. In 
August 1918 Ahmad al-Sharif left Tripolitania aboard 
a German (or Austrian) submarine and abandoned 
the supervision of the tarika. 

The subsequent period reflects the political history 
of Libya. After a long period of ambiguity in the rela- 
tions between Italians and Sanusis, and the signing of 
accords which were never properly implemented, the 
final struggle began in the late 1920s: Italian troops 
entered Djaghbub in February 1926 and Kufra in 
January 1931. One of the leaders of the brotherhood, 
c Umar al-Mukhtar, kept alive the last embers of 
Sanusi resistance until he was captured and publicly 
hanged by the Italians in September 1931. 

During the Second World War, a Sanusi force was 
raised to fight alongside the British. The liberation of 
Cyrenaica (1943) and British support for Muhammad 
Idris cleared the way for the inauguration of the 
Sanusi monarchy at the head of an independent Libya 
(24 December 1951). Like other MaghribI Sharifian 
lineages in other periods, the larika sanusiyya thus 
became, in circumstances of peril for the Muslims, a 
symbol of political legitimacy, but these new disposi- 
tions proved impossible to maintain or to extend. The 
seizure of power by the "Free Unionist Officers" 
under the leadership of Colonel Kadhdhafi (1 
September 1969) [see libiya. 3] led in Libya to a 
lasting "excision from history" of the brotherhood, 
which has come to at least temporary extinction. 

Bibliography: C.C. Adams, The Sanusis, in MW 
(1946), 21-45; G. Albergoni, Variations italiennes sur 
un theme francais: la Sanusiyya, in CRESM, Con- 
naissance du Maghreb. Sciences sociales el colonisation, 
Paris-Aix 1984; G. Ciammaichella, Libyens el Fran- 
cais au Tchad (1897-1914). La confrerie se'noussie el le 
commerce transsaharien, Paris-Marseilles 1987; D.D. 
Cordell, Eastern Libya, Wadai and the Sanusiyya: a 
Tanqa and a trade route, in Jnal. of African History 
(1977), 21-36; A.S. al-Dadjdjanl, al-Haraka al- 
sanusiyya. Nash ^aluhd wa-numuwwuha fi 'l-karn al-tasi'- 
'■ashar, Beirut 1967; H. Duveyrier, La confrerie 
musulmane de Sidi Mohammad ben AH Es-Senousi et son 
domaine ge'ographique en I'annee 1300 de I'Hegire = 1883 
de notre ere, Paris 1884 (the classic form of the "black 
legend"); E.E. Evans-Pritchard, The Sanusi of 
Cyrenaica, Oxford 1949; H. Klopfer, Aspekte der 
Bewegung des Muhammad Ben c Ali As-Sanusi, 
Wiesbaden 1967; A. Martel, Aux origines de I'Etat li- 
byen: La Porte el la Sanusiya au Sahara, 1835-1922, in 
CRESM, Enjeux sahariens, Paris-Aix 1984, 233-9; 
B.G. Martin, The Sanusi brotherhood in Libya and the 
Sahara, in Muslim brotherhoods in 19th century Africa, 
Cambridge 1976; C.A. Nallino, Senussi, in En- 
ciclopedia Italiana, 1936, 395-7; idem, Le doctrine del 

fondatore della confraternitd senussita, in Raccolla di scritti 
editi e inedite, ii, Rome 1940, 395-410; J. -L. Triaud, 
Tchad 1901-1902. Une guerre franco-libyenne oublie'e? 
Une confrerie musulmane, la Sanusiyya, face a la France, 
Paris 1988; idem, La legende noire de la Sanusiyya. Une 

Paris-Aix 1995, 2 vols.; K.S. Vikor, Sufi and scholar 
of the desert edge. Muhammad b. AU al-Sanusi and his 
brotherhood, London 1995; N.A. Ziadeh, Sanusiyah: a 
study of a revivalist movement in Islam, Leiden 1958. 

(J.-L. Triaud) 
SAR-I PUL, "the head of the bridge", called by 
Arab geographers Ra's al-Kantara, is a town of 
Afghan Turkistan (lat. 36° 13' N., long. 65" 55' E., 
alt. 610 m/2,007 feet), on the Ab-i Safid, from the 
bridge over which it takes its name. It is not to be con- 
fused with a village near Samarkand or a quarter of 
NIshapur, both of the same name, each of which is 
historically as important as the Afghan town. Between 
the northern spurs of the Paropamisus and the sands 
to the south of the Oxus, in a fertile tract well-watered 
by streams from the mountains, but proverbially 
unhealthy, lay four Ozbeg khanates or petty prin- 
cipalities, Akca, Shibarghan, Maymana and Sar-i pul 
with AndkhuT (Andkhud), the independence of which 
was destroyed by the Durrani and Barakzay Amirs of 
Afghanistan in the mid-19th century. Of these prin- 
cipalities, Sar-i pul was the last to succumb to the 
ruler of Kabul. In 1865 the troops stationed there 
revolted against the Amir Shir C A1I, but the mutiny 
was suppressed by c Abd al-Rahman Khan, who even- 
tually succeeded as Amir; not long afterwards Sar-i 
pul lost the last vestiges of its independence, but the 
former geographical and political divisions of the prin- 
cipalities were preserved and their Uzbeg inhabitants 
exempt from liability to military service. 

The site of Sar-i pul is probably identical with that 
of the town Anbar or Anbir, one of the main centres 
of the mediaeval Islamic principality of Djuzdjan 
[q.v., and see EIr art Anbir (C.E. Bosworth)]. The 
modern town of Sar-i pul comes within the present 
Djuzdjan province of Afghanistan; in ca. 1955 it had 
an estimated -population of 5,000. 

Bibliography: Le Strange, Lands of the eastern 
Caliphate, 426; Hududal- C alam, tr. Minorsky, comm. 
335; J. Humlum, La geographic dAfghanislan, etude 
d'un pays aride, Copenhagen 1959, 132, 151. 

(T.W. Haig*) 
SARA, wife of the Biblical patriarch 
Abraham [see ibrahim]. Sarah enters the text of the 
Kur 3 an only in its rendition of the etiological nar- 
rative surrounding the name Isaac (Hebrew wayyishak 
"and he laughed") from Gen. xvii, 15-22, xviii, 11- 
15 and xxi, 5-7; thus in Kur'an, XI, 71-3, and LI, 29- 
30, Sarah laughs at the messengers who bring the 
news that she will bear a son in her (and Abraham's) 
old age, but she remains unnamed and is referred to 
simply as imra-'aluhu "his [Abraham's] wife". The 
issue of Sarah's laughing (and thus doubting God) 
drew some Muslim exegetical attention as it did in the 
Jewish tradition (e.g. al-Tabari, Tafsir, ad XI, 71); the 
solution cited by al-Farra 3 , Ma c dnt 'l-KuPan, Cairo 
1955-72, ii, 22, and Ibn Kutayba, Gharib al-Kur'an, 
ed. A. Sakr, Cairo 1958, 205-6, and then in many 
sources after them, avoids the problem by defining the 
verb dahika as "menstruate" rather than "laugh" 
(justified by the idea that "rabbits laugh when they 
menstruate"), reflecting a popular rabbinic gloss of 
the Hebrew WnaA in Gen. xviii, 12, as "menstruate" 
rather than "pleasure" (see M.M. Kasher, in En- 
cyclopedia of Biblical interpretation, New York 1953, iii, 


In the wider Muslim elaboration of Biblical history, 
Sarah plays a substantial role. The story of her beauty 
and her being taken by Pharaoh, leading to the pro- 
clamation by Abraham that "She is my sister", 
created a great deal of discussion (see R. Firestone, 
Difficulties in keeping a beautiful wife: the legend of Abraham 
and Sarah in Jewish and Islamic tradition, in Jnal. of Jewish 
Studies, xlii [1991], 196-214). The incident became a 
part of an early tradition concerning the "three lies of 
Abraham" (e.g. Abu c Ubayd, al-Khutab wa 'l- 
mawaHz, ed. Ramadan c Abd al-Tawwab, Cairo 1986, 
1 13) as well as creating extensive discussions concern- 
ing the true lineage of Sarah (see Firestone, The 
problem of Sarah 's identity in Islamic exegetical tradition, in 
MW, lxxx [1990], 65-71; idem, Prophethood, mar- 
riageable consanguinity, and text: the problem of Abraham and 
Sarah's kinship relationship and the response of Jewish and 
Islamic exegesis, in JQR, lxxxiii [1993], 331-47). One 
tradition also states that her original name was 
Yasara, but this was changed to Sara when she was 
promised Isaac, and the yP was given to John (the 
Baptist) whose name was changed, through this addi- 
tion, from Haya to Yahya (see Abu Rifa c a al-FarisT, 
Bad^al-khalk wa-kisas al-anbiya\ in R.G. Khoury (ed.), 
Les legendes prophetiques dans I'Islam, Wiesbaden 1978, 
308). This may be compared with the change from 
Sarai to Sarah in Gen. xvii, 15, and the rabbinic com- 
ments on the status of the ^W in her name (see Kasher, 
ii, 248, where reference is made to the change of name 
from Hoshea to Joshua (Num. xiii, 16) with the addi- 
tion of a yod from Sarah's name is also cited). 

The sacrifice of Abraham's son [see ishak; isma c Tl] 
also brings Sarah into the accounts, especially as it ties 
into the understanding of her death. The concern of 
the son for the fate of his shirt and whether this would 
cause his mother grief (and lead to her death) is prom- 
inent. Also, Sarah is portrayed as a true believer in 
God alongside Abraham and her son, as within 
Judaism and Christianity (see Hebr. xi, 11), in the ac- 
counts of her resisting Satan's temptation to interfere 
in the sacrifice after Satan tells her about Abraham's 
plan (see Abu c Ubayd, al-Khutab, 111-12; al-Tabari, 
i, 293). 

Bibliography: Given in the article; see also J. 

Horovitz, Koranische Untersuchungen, Berlin-Leipzig 

1926, 114, on the pre-Islamic use of the name; R. 

Firestone, Journeys in Holy Lands, Albany, N.Y. 

1990._ (A. Rippin) 

SARA [see saray]. 

SARAB (a.), mirage. 

ral phenomenon. Sarah is specifical- 

e illus 

n of water (so 

to a sense of the verb saraba) seen at midday which ap- 
pears to be on the ground, as compared to al, which 
is seen early and late in the day and makes things ap- 
pear to float in mid-air and quiver. The lex- 
icographical tradition attempts several ways of 
distinguishing these two words, but the emphasis on 
the time of day of their appearance is the most consis- 
tent differentiation. Sarab is used twice in the Kurgan, 
in XXIV, 39, within a simile for the deeds of the 
unbelievers, the value of which turns out to be a 
mirage before God, and in LXXVII, 20, in reference 
to the last day on which mountains will vanish (into 
thin air), as in the experience of a mirage. 

The image of a "mirage" as reflected in Kur'an, 
XXIV, 39, gains powerful resonances in Suli 
thought, especially Ibn c Arabi, Futuhdt Makkiyya, 
Bulak 1329/1911, ii, 105, where the verse is inter- 
preted to suggest that God appears to people in the 
form of their need; pursuing that form, which is a 
mirage, reveals the non-being of God and leads to the 

Bibliography: EP , s.v. al; Lane, i, 127-8 (s.v. 
al); tafsir and Kur'an-lexicographical tradition esp. 
on XXIV, 39; W.C. Chittick, The Sufi path of 
knowledge. Ibn al-'-Arabi's metaphysics of imagination, 
Albany, N.Y. 1989, 378-9. (A. Rippin) 

2. In poetry. The mirage or fata Morgana, in its 
two forms distinguished by the Arabs (sarab and al, see 
above), is very often described by pre- and early 
Islamic Arabic poets in brief passages rarely ex- 
ceeding two lines, incorporated in their kasidas con- 
taining desert scenes. Often it is said to move, dance 
or "amble" (dfard, rakasa, khabba), to glitter or shim- 
mer (tarakraka, lama'-a) like water, enveloping hills and 
sand dunes that appear clothed in it, or partly hiding 
camels that seem to wade or move in it like ships. Its 
apparent movements make it seem alive, like animals 
or humans. Later, urban, poets remain fond of the 
images, the main differences being that the mirage, 
now become part of the poetic universe rather than an 
everyday reality, generally changes from a primum in- 
to a secundum comparationis, and that it is increasingly 
used less as a potential source of strongly visual im- 
agery than as a symbol of deceit, thwarted expecta- 
tions, illusions and unfulfilled promises, to point a 
moral rather than adorn a scene. The frequent occur- 
rence of this image is no doubt partly due to the 
Kurgan, where the sarab already stands for illusion and 
futility (XXIV, 39) or instability (LXXVIII, 20). For 
small selections of the vast but scattered material 
(Dhu '1-Rumma's Diwan alone offers over thirty 
passages), see e.g. Ibn Abi c Awn, Tashbihat, London 
1950, 71-4; Abu '1-HilaI al- c Askari, Diwan al-ma'tni, 
Cairo 1352/1933-4, ii, 128-9; ShimshatI, al-Anwir wa- 
mahastn al-ashfar, i, Kuwait 1977, 355-9. 

(G.J.H. van Gelder) 
SARACENS, a vague term used in the West for 
the Arabs and, eventually, other Islamic peoples of 
the Near East, in both pre-Islamic and mediaeval 

1. Earlier usage. 

Saracens was one of the many terms that Classical 
authors and ecclesiastical writers, used for the Arabs, 
the others being Arabes, Skenitai, Tayyaye, Ismailitai 
and Hagarenoi. It became the most common of all 
these terms, although it was one that the Arabs did not 
use in referring to themselves. The term was a coinage 
composed of 'Sarak and the Greek suffix enos, and 

Many etyma have been suggested for this term, such 
as shark (the east), saraka (steal) and Sawdrik (a tribe), 
but all of them have been weighed and found wanting. 
The most recent effort, one that has received a wide 
vogue, considers the etymon to be shrkt, federation, a 
term to be found in the bilingual Greek-Aramaic in- 
scription of Thamud in Ruwafa. Many serious objec- 
tions were advanced against this etymology, even 
before a close examination of this inscription revealed 
that the term shrkt was a misreading of shrbt (tribe), a 
conclusion supported by the Greek version of the in- 
scription which has ethnos, the exact equivalent of 

The failure of all these attempts to explain Sarakenos 
calls for a return to what the Classical authors said 
about the term, accepted long ago by Noldeke as the 
true etymology. The earliest certain attestations of it 
go back to the 1st century A.D. when Dioscorides of 
Anazarbos, a physician-pharmaeologist, spoke of the 
Saracen tree imported through the Nabataeans, and 
Pliny the Elder, who spoke of the Saracens as the Ar- 
cani who lived beyond the Nabataeans. Thus the two 
authors pinpointed the Saracens as a people in north- 
western Arabia and clearly indicated that the term 
was not a generic but specific one. Ptolemy in his 


Geography speaks of a district Sarakene in Arabia 
Petraea, and Stephanus of Byzantium in his Ethnika 
speaks of Saraka as a district beyond the Nabataeans 
whose inhabitants are called the Sarakenoi; although a 
6th century writer, Stephanus depended on the much 
earlier works of Ulpianus and Uranios. These writers 
provide data for the most plausible etymology of the 
term, that relating to a region, Saraka. This is sup- 
ported by the fact (unnoted by the etymologists of this 
term) that the Greek suffix enos is used to form ethnic 
adjectives from geographical names. 

Surprisingly, however, Saraka seems to have disap- 
peared from Arabian toponymy, although Arab 
geographers know of two almost homophonous place- 
names in the Hijaz, a valley (Suwarik) and a village 
(al-Suwarikiyya). Serious consideration should, there- 
fore, be given to the possibility that Saraka is none 
other than Sara(t), the well-known mountain range in 
that area. It is either a dialectal version of Sara(t) or 
it experienced epenthesis by the intrusion of the kappa. 
Even more important is the denotation of the term . 
It clearly was applied originally to a group of Arab 
pastoralists in northwest Arabia, but soon it became 
the generic term for all the Arab pastoralists within as 
well as without the Roman limes. And so it was used 
by the secular and ecclesiastical historians of the 4th 
century, such as Ammianus Marcellinus and 
Eusebius respectively. The latter developed an in- 
terest in both the term and what the term stood for. 
For them, the Saracens were the Biblical people, the 
sons of Ishmael, hence the children of the bondwoman 
Hagar and thus "outside the promises". As some of 
the Saracens attacked the inmates of the monastic 
establishments in the Orient, such as those in 
Chalcidice and the Desert of Judah near the Jordan, 
they acquired a bad reputation among these ec- 
clesiastical historians, some of whom indulged in 
etymologising the term "Saracen" along pejorative 
lines. Such was Sozomen, who suggested that in order 
to avoid the opprobrium attaching to their descent 
from Hagar, the bondwoman, the Arabs started to 
call themselves by a name that related them to Sarah, 
the wife of the first patriarch. These perceptions of the 
Arabs most probably explain the emergence of other 
Biblical terms for the Arabs, such as Ismailitai and 
Hagarenoi, with the same pejorative implications. 

Latin authors who came to the East, such as 
Jerome, also etymologised the term along these 
Biblical lines and so projected an uncomplimentary 
image of the Saracens. Jerome's unfortunate ex- 
perience with some marauding Saracens while he liv- 
ed in Chalcidice and at Bethlehem, contributed fur- 
ther to the deterioration of that image in his writings. 
Such was also the experience of the pilgrim, the 
Anonymous of Placentia, with the Saracens of Sinai in 
the 6th century. Thus through the writings of such in- 
fluential figures as Jerome, the term "Saracen" with 
all its pejorative implications reached Western Europe 
long before the Muslim Arabs appeared in the Roman 
Occident in the 2nd/8th century. 

Bibliography: Dioscorides, De Materia Medica, 
ed. M. Wellmann, Berlin 1958, i, 60; Ptolemy, 
Geography, ed. C.F.A. Nobbe, Hildesheim 1966, 
book V, ch. 17, p. 69; Stephanus of Byzantium, 
Ethnica, ed. A. Meineke, Berlin 1849, 556; Yakut, 
Mu^djam al-bulddn, Beirut 1957, iii, 275-6; Am- 
mianus Marcellinus, Res Gestae, XXII. 15.2; 
XXIII. 6. 13; Sozomen, Ecclesiastical history, ed. J. 
Bidez, Berlin 1960, book VI, ch. 38; D.F. Graf and 
M. O'Connor, The origin of the term Saracen and the 
Rawwafa inscriptions, in Byzantine Studies/ Etudes byzan- 
tines, iv (1977), 52-66; O'Connor, The etymology of 
Saracen in Aramaic and pre-Islamic contexts, in The 

defence of the Roman and Byzantine East, BAR, Interna- 
tional Series, 297, ii (1986), 603-32; I. Shahid, 
Rome and the Arabs, Washington D.C. 1984, 123-41; 
idem, Byzantium and the Arabs in the fifth century, 
Washington D.C. 1989, 484, 485, 543; B. Moritz's 
art., Saraka in Pauly-Wissowa and J.H. Mordt- 
mann's Saracens in the EP may be consulted for the 
old controversies on the etymology of Saracenus. 

!. In n 

il Europea 


t Shahid) 

: Saracen" (OFr. Sar(r)azin, Sar(r)acin, 
OEng. Sarracene), whatever its origin (see above, 1.) 
came into both Late Greek and Late Latin usage 
during late antiquity, and at that time simply meant 
"Arab". With the rise of Islam, and in subsequent 
mediaeval European times, European writers used 
"Saracen" to denote "Arab" or "Muslim" or both, 
according to context. With the contacts of the 
Crusaders with the Saldjuks of Rum in the 6th/12th 
century and the rise of the appellation Turcia for Asia 
Minor, and, above all, with the appearance of the Ot- 
tomans, "Saracen" in the sense of "Muslim" gave 
place to "Turk". As the Christian reconquista in the 
Iberian peninsula progressed, followed by Spanish 
and Portuguese attacks on the North African 
mainland, "Saracen" in the sense of "Arab" began 
to be generally replaced by "Moor", since the Christ- 
ian peoples of the Iberian peninsula had used Mauri 
and Moros for the Arab-Berber invaders of their land 
[see moors]. The increasing numbers of Western 
travellers in North Africa and the Near East also came 
to use "Arab" more particularly in a pejorative sense 
(here following the usage of the indigenous urban 
populations of those lands) for "Bedouins, brigands". 
Hence in Western Europe, by the later Middle Ages, 
"Saracen" had tended to fall out of usage and to be 
replaced by somewhat more specific terms. 

The great literary usage of "Saracen" in the Euro- 
pean high Middle Ages was, of course, to designate 
the Muslim opponents of the Christians in the Chan- 
sons de geste, mostly set fictiously in the time of 
Charlemagne or his son Louis I ("The Pious") and 
written in the various forms of Old French. From 
these originals, whose genesis in time is uncertain, the 
views and concepts of the Chansons spread to other ver- 
nacular literatures, such as Spanish, Provencal and 
Italian, and also to English and German literature. 
These Chansons provide for us an idea of what was the 
unofficial, un-clerical, un-academic view of the 
Muslims (as opposed to the views of monkish and 
other clerical polemicists) during the Middle Ages. 
But whether their (largely unknown) poetic creators 
meant to provide a realistic portrait, as they conceived 
it, of their Muslim foes in Spain, North Africa, Syria 
or Egypt, or whether they used "Saracen" in a vague, 
general sense, is not always clear. 

Byzantine usage of the term "Saracens" probably 
lasted till the end of the empire. Certainly, Ibn Bat- 
tuta was addressed as such (kul li-hddhd 'l-Sarakinu ya'-ni 
al-Muslim) by a monk in Constantinople (H.A.R. 
Gibb pointed out that it was chronologically impos- 
sible, however, that this monk could have been the ex- 
Emperor Andronicus II) when he was there, probably 
in the autumn of 732-3/1332 (Rihla, ii, 441-2, tr. 
Gibb, ii, 512-13). 

Bibliography: The literature on Christian con- 
ceptions of Muslims as shown by the chansons degesle 
is extensive. A useful start can be made from the de- 
tailed bibliographical information in N. Daniel, 
Heroes and Saracens. An interpretation of the Chansons 
de Geste, Edinburgh 1984, 280-1 (studies), 320-7 
(texts). (C.E. Bosworth) 

SARAJEVO, principal city of Bosnia (and of 

the republic of Bosnia-Herzegovina of former 
Yugoslavia), situated on the banks of a small river 
known as the Miljacka (western tributary of the river 
Bosna), at the outlet of a valley opening towards the 
West, this being the "Plain of the Seraglio"— in 
Turkish Saray Ova or Saray Ovasi (hence the name of 
the city), surrounded by tall and precipitous moun- 
tains (notably those of Ozren (1 ,452 m/4,762 feet) and 
of Trebevic (1,629 m/5,343 ft)), at an altitude, ac- 
cording to the various neighbourhoods, of between 
537 m/1,761 ft and 700 m/2,296 ft. 

Its first mention, under the name of Vrhbosna ("the 
Crete of Bosnia", of which the zupa (local parochial 
district) of the same name is mentioned from the 10th 
century onwards) apparently dates from 1415, but 
this appellation in fact only applied to the fortress 
situated on a rocky promontory dominating the city 
(and the small township in front of the latter), since 
the city of Sarajevo as such, for a long time called 
simply Saray (meaning the seraglio, i.e. the 
"palace"), or Bosna Saray, was founded ca. 1429 ac- 
cording to some authors, but quite certainly at a later 
date, by the Ottomans (see e.g. B. Djurdjev, art. 
bosna, at vol. I, 1263a). 

The fertile region around Sarajevo, with its abun- 
dant water sources and forests, was inhabited from the 
Neolithic period (2400 to 2000 B.C.), as is proved by 
the excavations of Butmir (one of the principal urban 
centres of the Balkan peninsula in this period, the 
ceramics of which are renowned). Other prehistoric 
dwellings have been brought to light on the slopes of 
Trebevic (as on those of Debelo Brdo), some of which 
may have possibly existed even before the Roman 
period. Towards the end of the Bronze Age (900 
B.C.), this region experienced an Illyrian influx, of 
which numerous vestiges have been discovered. In the 
Roman period, the 8th Augustan Legion was based 
on the plain of Sarajevo, and the well-known 
sulphurous bathing establishments of Ilidza, a ther- 
mal station situated in the foothills of Mount Igman 
(today some ten kilometres from the centre of the city) 
were developed. Other Roman remains have been 
discovered within Sarajevo itself. 

After the fall of the Roman Empire, the Eastern 
Goths were the first to establish themselves in this ter- 
ritory, then, in 535 A.D. the entire region was con- 
quered by the Byzantine Emperor Justinian I. During 
the 7th century Slavic tribes occupied the region, and 
from the 10th century onwards there began to be 
formed, in the depression between the current cities of 
Sarajevo and of Zenica, the nucleus of the future 
mediaeval Bosnian state. It is known, on the other 
hand, according to a document of the king of 
Hungary Bela IV dating from 1244, that at that time 
the territory of Sarajevo formed a part of the zupa 
(local parochial district) of Vrhbosna, then the site of 
the cathedral of Saint Peter, centre of the diocese. As 
early as 1379, the presence was noted at Vrhbosna of 
traders from Ragusa, who mentioned a locality called 
Trgoviste (meaning "the place of [open] market") 
situated in the territory of what is now the city of Sara- 
jevo, at the point where the stream known as Kosevo 
(currently the name of a quarter of the city) joins the 
river Miljacka. It is also known that in 1415 a local 
dignitary, the voi'vode Pavle Radenovic, was buried at 

The first decades of the 15th century were marked 
by increasingly frequent incursions on the part of the 
Ottoman cavalry, who took possession in 1416 (or 
possibly not until 1428) of the fortress of Hodidjed, a 
strategic position commanding the valley of Miljacka, 
situated two hours' march from Vrhbosna. Hodidjed 

was definitively captured by the Ottomans in the sum- 
mer of 1435 (according to some sources in the 
previous year), thus some thirty years before the lay- 
ing of the first foundations of the future city of Sara- 
jevo, intended as a secure base for the conquest of 
Bosnia, of Herzegovina, of northern Serbia, of 
Dalmatia and of a part of Croatia, then also, a century 
later, of a part of Hungary. 

The Ottomans recognised at a very early stage the 
value of its location, and when they conquered Bosnia 
during the time of Mehemmed I, in the spring of 
867/1463, they made it the principal arsenal of the 
conquered territory. From 1438-9 onward an Ot- 
toman governor was installed there, with the duty of 
controlling the indigenous local dynasties (in par- 
ticular that of the Pavlovics) who were required to pay 
tribute. After the definitive conquest of the kingdom 
of Bosnia, and the execution of its last king, Stjepan 
Tomasevic, the Ottoman governor at first resided at 
Vrhbosna, the name of which was to be retained for 
some time (at least until the beginning of the 16th cen- 
tury), as is shown by the travel-writing of Felix Petan- 
cius and of Benedict Kuripesic, and exchanges of let- 
ters with Ragusa (where the following forms are 
found: Werchbossen, Verchbossen, Verbosavia, Ver- 
bosania, Verchbossania, etc.). 

The first significant foundations date from the years 
1460-1 (c.'. the wakf-name of 1462 of c Isa Beg, son of 
Ishak Beg, who was to become sandjak beg of Bosnia in 
1464): initially, the governor's palace and a wooden 
mosque, then a bridge over the Miljacka, a 
caravanserai, a bedesten, a hammam, residential houses, 
shops, water-mills, etc. It is said that from 1455 on- 
ward the new urban centre (kasaba), was called some- 
times Saray, sometimes Saray owasi, or even Saray 
kasabasi, the first mention of its current form (which is 
found in a letter written in Cyrillic characters) dating 
from 1507. Twenty years after the first constructions, 
in 1480, the city was taken by assault and burnt, in 
the course of a raid mounted jointly by the Hungarian 
garrison of Jajce led by Peter Doczy, and Serbian 
troops under the command of the despot Vuk 
Grgurovic/Brankovic ("Zmaj Ognjeni Vuk"). 

Sarajevo's most illustrious period belongs to the 
16th-17th centuries. It corresponds, of course, to the 
era of the greatest expansion of the Ottoman Empire 
in the Balkans and in South-Eastern Europe, but also 
to the Ottoman golden age per se, a period in which the 
massive quantities of booty amassed in the course of 
against ' 

brought prosperity t 

o commerce and to craftsman- 

n the development of a large 

number of towns. Sa 

ajevo was for close on a century 

the s< 

1463 t 

>f the g 


which c 

e the 


transferred to Banja Luka [q. v. ] before being restored 
to Sarajevo in 1637-8). Its governors ( c Isa Beg, Ayas 
Beg, YahyS Pasha, Iskender Pasha (whose name is 
still born by the quarter of "Skenderija"), then the 
son of the latter, Mustay Beg, and the most illustrious 
of them, Ghazi Khosrew Beg, a native of the town of 
Trebinje in Herzegovina, who lived for seventeen 
years on the banks of the Miljacka (he was several 
times sandjak beg of Bosnia between 1521 and the year 
of his death, 1541, and was buried in Sarajevo), 
embellished it with the construction of a large number 
of renowned buildings: the mosques of c Is5 Beg 
(926/1520), of GhazT Khosrew Beg (937/1530 [see 
;~ 'ith a medrese built in 1537, a library, 
ie, a public kitchen, a hospice for travellers, a 
-built khan, a bedesten, a hammam, etc.), of Gh azi 
Pasha (969/1561), of Ferhad Beg (also in 

:he Imperial Mosque 
1566 at the order of 

to replace the former 
founded in 1457 and 
others; the tekkes of the 
that of the 

969/1561), not forgetting 

"Careva dzamija" (built i 

Siileyman the Magnificent. 

mosque of the same name 

destroyed in 1480), and mar 

MewlewTs, of the Khalwetl: 

Kadiris (the renowned tekke of Hadjdji Sinan Agh; 

constructed in 1638-40 and subsequently restored on 

numerous occasions); the bedeslens (in particular the 

well-known "Bursa bezistan" built by Riistem Pasha 

[q.a.] in 1551); the clock tower (built at the end of the 

16th century or at the start of the 17th), the medreses, 

the tiirbes, the fountains, the baths, the khans, etc. 

The number of inhabitants of the city, which had 
gained the status of shehir before the 16th century, 
grew rapidly as a result of the influx of the Muslim 
population, which settled at the outset on the left bank 
of the Miljacka; for a very long time, each religious 
group lived in separate mahalles. This population con- 
sisted above all of new converts — there had been pro- 
gressive Islamisation of a significant section of the 
local Slav population — as well as numerous ad- 
ministrative and religious cadres, Ottoman civilians 
and soldiers, of diverse origins and belonging to the 
most varied ethnic groups, as is demonstrated by the 
genealogies of some of the eminent Muslim families of 
the city. But the city also expanded as a result of an 
influx of indigenous Christian populations. It is in- 
teresting to note in this context that in 1477 there were 
in Sarajevo 103 Christian households, 8 households of 
Ragusans, and only 42 Muslim households (see Ham- 
dija Kresevljakovic, Esnafi, 1958, 9, quoting an article 
of Nedim Filipovic). The Christian population was 
composed of the Orthodox — whose Old Church, 
"Stara Crkva", was built in 1528, then rebuilt on 
several occasions after numerous fires, in particular in 
1616 and in 1658, subsequently reconstructed com- 
pletely in 1730, then once more renovated in 1793 
(the list of popes of Sarajevo from 1516 to 1804 may 
be found in V. Skaric, Srpski..., 140-1)— and of 
Catholics, some of whom came from Ragusa [q.v.] in 
the course of the second half of the 16th century and 
settled in a separate quarter, subsequently called 
"Latinluk". Not to be ignored is the arrival, also 
around the middle of the 16th century, of a relatively 
substantial Jewish colony. These were, of course, 
Jews expelled from Spain and Portugal, who settled in 
a quarter later known as "Cifuthana". The 
synagogue was probably built around 1580, then com- 
pletely renovated in 1821 , having been twice damaged 
by fire, in 1697 and in 1788. The total number of 
hearths (in Turkish odjak) thus apparently increased 
from 153 in 1477, to 181 in 1480, subsequently to 
1024 in the first half of the 16th century, then to 4270 
in the second half of the 16th century. Aided by 
geographical position and "the industries of war", 
commerce and craftsmanship developed rapidly, as is 
clearly shown by the number of warehouses and 
covered markets, of traders and of types of merchan- 
dise which were sold there or which passed through 
the city. In fact, the city was located "on the caravan- 
route leading from Istanbul and from Salonica 
towards the West, at a staging-point where it was 
necessary to substitute horses and mules for camels" 
(G. Veinstein, op. tit., in Bibl., 92). It was linked to 
the Adriatic coast on the one hand by the valley of the 
Neretva, on the other by the route situated further to 
the north-west, leading from Livno to Split. Further- 
more, in the local context, the city, situated at a cross- 
roads, was also an excellent outlet for agricultural 
markets. In a totally different domain, it may be 
added that in 1085/1674-5 under Mehemmed IV and 

in 1099/1687-8 under Suleyman II [q.vv.], copper 
coinage was struck there (these being the coins known 
as mankiT). This economic prosperity was naturally ac- 
companied by increasingly intense religious and 
cultural activity, for Sarajevo had very rapidly 
become an important administrative centre. With 
reference to the Muslim population, there is abundant 
testimony (cf. for example the works cited in the Bibl. 
of H. TahmisHc, of M. Mujezinovic and of H. 
Sabanovic); in the literary domain as such, the best 
known names for the whole of this period remain 
those of Mehmed Nergisi(d. 1044/1635 [q.v.]) and of 
Hasan Ka'imi (d. ca. 1101/1690, cf. J. Samic, op. 
eit.). It should, however, be noted that the overall 
development of the city was thwarted on numerous 
occasions by various scourges: outbreaks of plague 
(like that of 1526-7) and of cholera (1691), fires 
(particularly worth mentioning are those of 1644 and 
1656) and earthquakes, not to mention famines. In 
spite of all this, the description of Sarajevo in 1660 
provided by Ewliya Celebi (even bearing in mind the 
exaggerations characteristic of this author) is quite im- 
pressive: the city reportedly then comprised 400 
mahalles, including ten Christian ones and two Jewish 
ones (it may be recalled that according to an earlier 
source at the end of the 16th century, the city allegedly 
comprised 91 Muslim mahalles and two Christian 
mahalles), 17,000 houses, 77 mosques and 100 mesdjids, 
a clock tower, numerous medreses and other specialised 
religious schools, 180 mektebs, 47 tekkes, 110 public 
fountains, 300 sebils, 700 wells, 76 flour mills, five 
hammams, 670 private bathrooms, three caravan- 
serais, 23 khans, 1,080 shops, a bedesten, seven bridges 
over the Miljacka, an Orthodox church, a Catholic 
church, a synagogue, seven c ima~rets, etc. 

This long period of prosperity was brusquely inter- 
rupted four decades later, in September 1697, by a 
terrible and totally unexpected blow, the sacking and 
burning of the city by Austrian troops commanded by 
Prince Eugene of Savoy who, taking advantage of his 
victory over the Ottomans before Zenta, carried out 
an exceedingly bloodthirsty raid, leaving behind him, 
after a brief occupation of the city, the ruins of Sara- 
jevo ablaze. The unsuccessful siege of Vienna attemp- 
ted by the Ottomans in 1683 marked, effectively, the 
beginning of a totally different period, that of the 
reconquista, and of the definitive withdrawal of Ot- 
toman troops from Hungary and from Slavonia, but 
also from Vo'ivodina, from Croatia and from 
Dalmatia. A new era also began for the city of Sara- 
jevo (the seat of the Ottoman wezirs was furthermore 
transferred after 1699 to Travnik, where it remained 
until 1850), a period during which relations between 
the Muslim and non-Muslim populations of the 
region deteriorated sharply, as did relations between 
on the one hand the ruling classes composed of in- 
digenous Muslims, the aghas, the a c ydns and the local 
begs, struggling fiercely to preserve their long-standing 
privileges, and on the other the Porte, resolutely pro- 
moting a whole series of new reforms — for the most 
part exceedingly unpopular— and its representatives 
who, appointed for very short periods, sought to 
enrich themselves with maximum haste at the expense 
of the indigenous populations, irrespective of religion, 
although the non-Muslims bore the heaviest burden. 

In the 18th century, the economic and financial 
crisis of the Ottoman Empire, following the crisis of 
the very structures of the state, considerably weaken- 
ed the latter's military power. This had immediate 
repercussions for the whole of the eyalet of Bosnia, 
henceforward a frontier region bordering on Christian 
Europe, as well as for the city of Sarajevo, where 

disorder and corruption became rife. In fact, the ar- 
rival in the city of huge numbers of Janissaries forced 
to leave the vast territories conceded to the "infidels" 
provided a ready source of troops for rebel governors, 
who relied upon them on every occasion, also upon a 
large number of malcontents among the aggrieved 
local sipdhis and the Muslim masses of the city, whose 
standard of living had worsened considerably since 
the beginning of the reconquest, with the constant in- 
crease in levies and the creation of new taxes. There 
ensued a series of revolts and seditious activities, 
punctuated by full-scale internal wars such as those 
conducted, on behalf of the central power, after the 
major revolt against the Porte which erupted in Sara- 
jevo in 1750, by Mehmed Pasha Kukavica, a native 
of Foca in Bosnia (from 1752 onwards), or by the kul 
cawuI'AK Agha (from 1772 onwards). These difficult 
times were accompanied by a whole series of scourges, 
first plague, which raged on a number of occasions in 
the course of this century (in 1731-2, 1741, 1762-3 
and 1781-2), then numerous fires which devastated 
Sarajevo on some ten occasions (in 1721, 1724, 1731, 
1748, 1766, 1769, 1773, 1776, the most serious occur- 
ring in 1788 and 1797), as well as numerous floods. In 
spite of all this, efforts were made to restore certain 
ancient buildings, such as the fortress situated in the 
old town (in 1729-39), the mosque known as 
"Magribija" (constructed in the 16th century and en- 
tirely rebuilt in 1766), the Serbian Orthodox church 
(rebuilt in 1730 and fully restored in 1793— we are 
told that in 1720 the city reportedly contained be- 
tween 3,000 and 5,000 Serbs, cf. V. Skaric, Srpski, 54, 
but this number must also include the Serbs of the 
surrounding villages gravitating round the Orthodox 
church in Sarajevo), etc. It may be recalled finally 
that the history of Sarajevo in the second half of the 
18th century is drawn from an exceptional source, the 
Chronicle of Mulla Mustafa Baseski (1731-2 to 1809), 
which covers the years 1746-1804, and which contains 
a mine of first-hand information. 

At the beginning of the 19th century, general 
discontent and resistance to reforms continued in 
Sarajevo, just as in other regions of Bosnia and of 
Herzegovina, to the point where insurrections against 
the governors sent by the Porte (or even sometimes 
directly against the central power) were carried out 
with increasing intensity. This was particularly the 
case in 1814 and then in 1826 (date of the major revolt 
which followed the suppression of the corps of 
Janissaries). But these revolts were invariably brutally 
suppressed, for example, by Djelal ul-DIn Pasha in 
1829, and subsequently by c Abd ul-Rahman Pasha. 
However, soon afterwards, in 1246/1831, a new ma- 
jor uprising erupted, this time against reforms in the 
organisation of the Ottoman army and led by the 
kapudan Huseyin Beg Gradascevic, nicknamed "Zmaj 
od Bosne" (i.e. "Dragon of Bosnia"). The movement 
spread rapidly at first, both in Bosnia and 
Herzegovina, but when, some time later, other 
kapuddns dissociated themselves from it and joined the 
side of the government forces [see ridwan begovic], 
it too was suppressed, with much bloodshed, after a 
decisive battle which took place at Pale near Sarajevo. 
(Attention should be drawn in this context to a 
tendency which has been observed throughout the 
Balkans: during the Communist period, some 
historians and pseudo-historians of former Yugoslavia 
represented these various uprisings against the Porte, 
led by various local condottieri, as "national revolts" of 
the indigenous Muslim populations against the Ot- 
toman Turks, which is manifestly false.) In spite of 
these numerous setbacks, some years later (in' 1840 

and then in 1848), there was a renewal of uprisings 
against the wezirs of the Porte based in Travnik, which 
induced the latter to embark on a wide-ranging policy 
of repression. The task was entrusted, in 1850, to 
c Omer Pasha (formerly a junior officer in the 
Austrian army, a native of Lika, in Krajna, a region 
of Croatia, whose name before his conversion to Islam 
was Mihailo (Mica) Latas (1806-71)), who, armed 
with special powers and a substantial military force, 
definitively crushed all resistance in 1850-1, executing 
in the process a large proportion of the indigenous 
Muslim ruling class, just as he had done previously, 
in actions of a similar type, in Syria, in Albania and 
in Kurdistan, or was later to do in Montenegro and 
in Herzegovina, although his efforts in Crete in 1867 
were unsuccessful. After these bloody events, the seat 
of the Ottoman wezirs in Bosnia was definitively 
transferred from Travnik to Sarajevo. From this time 
onwards, the city experienced the implementation of 
a number of reforms aimed at European-style moder- 
nisation, as during the vizierate of Topal c Othman 
Pasha (i.e. between 1861 and 1869), a period which 
saw a hesitant and belated reform of education, the 
establishment of the first Ottoman printing-press in 
these regions (that of the wilayel), and thus the ap- 
pearance of the very first local Muslim journals (cf. 
Dj. Pejanovic, op. cil. , in Bibl.). Two other 
phenomena affected the city of Sarajevo substantially 
from the mid-19th century onwards: on the one hand, 
the gradual and final disappearance of the organisa- 
tion of the esndf (guilds of craftsmen) which had 
dominated the economic life of the city in preceding 
centuries; on the other hand, and most importantly 
(as this was to change enormously the relations ex- 
isting between the various populations of the city), the 
gradual but constant enrichment of many of the 
Serbian families of Sarajevo, who were subsequently 
to represent a considerable social force in the material 
and spiritual life of the city. It is, however, worth 
remembering that, in the words of a significant 
remark of M. Ekmefrc, "around the middle of the last 
century (i.e. the 19th), Sarajevo contained 100 mos- 
ques, and one Serbian Orthodox church" (cf. Srpski 
narod u Turskoj od sredine XIX veka do 1878, in Istorija 
srpskog naroda, v/1 [Belgrade 1981], 454). Finally, it 
may be noted that, as in the past, Sarajevo had to en- 
dure in the 19th century a prolonged outbreak of 
plague (in 1813-16), and a number of major fires (in 
1831, 1842, 1852, and the most devastating of all, in 

According to the resolutions of the Congress of 
Berlin (June-July 1878), Bosnia and Herzegovina 
were placed under the mandate of Austria-Hungary, 
although the latter did not formally annex the two 
provinces until October 1908. The troops of the Dual 
Monarchy entered the city of Sarajevo in August 
1878, having encountered a desperate resistance, as 
unexpected as it was murderous, on the part of a sec- 
tion of the Muslim population of the city, which lasted 
eight hours. It was led by numerous local individuals, 
including an imam, Hadzi Lojo (Loyo), although there 
was no significant involvement on the part of Sara- 
jevo's Muslim ruling class, which remained aloof 
from this popular movement. The forty years of 
Austro-Hungarian occupation (1878-1918) trans- 
formed the appearance of the city beyond recognition, 
not only in matters of town-planning and architec- 
ture, but also in terms of the religious (corresponding 
in this particular case to ethnic) composition of the 
population. This is clearly illustrated by the following 
table, devised by one of the two most knowledgeable 
historians of the city, Hamdija Kresevljakovic (the 

other specialist being Vladislav Skaric). What is estab- 
lished is on the one hand a quite spectacular fall in the 
percentage of the Muslim population, and on the 


:he Catholic population (see Kresevljakovic, 






































Seeing the occupation of Bosnia-Herzegovina as a 
first stage in its colonisation of the Balkans, the 
Austro-Hungarian government proceeded metho- 
dically towards the implementation of numerous pro- 
jects, especially in Sarajevo, designed to facilitate the 
attainment of this objective, while at the same time 
demonstrating to international opinion the civilising 
nature of its mission: construction of railway, of a cen- 
tral electricity system and of urban canals; improve- 
ment of the quays of the Miljacka; construction of the 
Catholic cathedral (1884-9, replacing the former 
Catholic church, on which little information is 
available, as is hardly surprising when the figures in 
the above table are considered); building of a 
monumental Town Hall in pseudo-Moorish style 
(1896), and many other public buildings on the grand 
scale (such as the magnificent Zemaljski Musej 
(1888), modelled on the Vienna Museum, the 
Theatre, the Law Courts, the Bank, the Protestant 
Church, the Hospital, schools, hotels, etc.; not forget- 
ting the laying-out, in 1886, of the city's first 
municipal park on the site of a Muslim cemetery). 
This rapid Europeanisation brought to Sarajevo many 
soldiers and officials. Among the latter, both in ad- 
ministration and in education, there were to be found 
a large number of Orthodox (i.e. Serbs) and Catholics 
(predominantly Croats, who were joined by con- 
siderable numbers of new arrivals from elsewhere: 
Austrians, Germans, Hungarians, Poles, Czechs, 
Slovaks, Slovenians, Rumanians, etc.) (see Kresevl- 
jakovic, op. cit., 37-9). The city was extended rapidly 
in a westwards direction, i.e. towards the plain, and 
towards Ilidza, which soon became a kind of "oriental 
Baden". At the same time, as was to be anticipated, 
there was a decline in the craftsmanship of the city, an 
inevitable consequence of the appearance of the first 

The shock experienced by the Muslim population 
of Sarajevo, as a result of Austro-Hungarian occupa- 
tion and the sudden irruption of all this modernity, 
was acute, as is shown by many texts of this period, 
and also by the emigration (although apparently of 
limited extent) of some of the inhabitants of the city 
to Turkey, or towards closer regions still controlled by 
the Ottoman Empire (see A. Popovic, Isl. balk., 272- 
3). However, gradually the Muslim reacted and 
organised themselves into a religious c 
guided by an Ulema medzlis and an administratis 
wakfs, at the head of which was the chief of the 

munity bearing the title of Re-'is al-ulema (see EP , I, 
1273b). In matters affecting schools, and education in 
general, great changes took place, since the Austro- 
Hungarian authorities completely reformed the 
organisation of public instruction. In this new system, 
which had little effect on the various Muslim elemen- 
■ schools, medreses, more or less "reformed", 
■ed for the training of religious functionaries of in- 
Dr status. The best-known in Sarajevo at this time 
e the Kursumlija and Hanika medreses, both dating 
n the time of Ghazi Khosrew Beg (cf. Spomenica 
i...). Furthermore, in 1887, a special college was 
igurated in Sarajevo with the aim of training 
judges for the Muslim courts and senior religious 
onaries. This was the highly-renowned Serialska 
Sudacka Skola u Sarajevu, which was the principal seed- 
bed of the Muslim religious intelligentsia of Bosnia- 
Herzegovina between 1887 and the Second World 
War, and w;here the duration of studies was five years. 
Many details concerning this institution are to be 
found in the two volumes compiled on the occasion of 
its thirty-fifth and fiftieth anniversaries: in 
Tridesetpetogodisnji izvjestaj Seriatske Sudacke Skole u Sara- 
jevu, Sarajevo 1917, and especially in Spomenica Seriat- 
ske Sudacke Skole u Sarajevu, izdana povodom pedeselgodisn- 
jice ovoga zavoda (1887-1937), Sarajevo 1937. In 1892 
an academy was also founded for the training of 
school-teachers (Dar al-mu c allimin), where the course 
of study lasted three years. It should be noted that 
there was at first, among the local Muslims, a period 
of passivity, of mistrust and of defiance regarding 
everything emanating from the Austro-Hungarian 
authorities. In this context, worth citing for example 
is the fact that in 1887, at the time of the inauguration 
of the Ser. Sud. Skola which has been mentioned above, 
the new administration encountered obstinate 
resistance on the part of the Muslims, who refused to 
send their sons to the school, with the result that the 
first pupils of this establishment were recruited among 
orphans (cf. Abduselam Balagija, Les musulmans 
yougoslaves, Algiers 1940, 115). But subsequently, as a 
result of a gradual transformation of opinions, many 
Muslim children began to attend secular elementary 
schools and academies. Some even pursued studies 
abroad, especially in Vienna and in Budapest, for 
those to whom Zagreb was not a preferable option, to 
qualify as doctors, engineers, etc. Others also went 
abroad, but with the object of pursuing traditional 
studies, in Istanbul, in Cairo and in Medina, or on 

the contrary, to become initiated into Western-style 
Islamology, in Vienna (as was the case with Sukrija 
Alagic, Fehim Bajraktarevic and Safvet-beg Basagic) 
or in Budapest (in the case of Sacir Sikiric), thus 
becoming the very first local orientalists. 

There is much that should be said regarding the 
Muslim press of Sarajevo during these four decades. 
This evidently reflects accurately the principal 
political, social, cultural and other tendencies which 
emerged in the Muslim community, a community 
which found itself from day to day, without really 
understanding how, being carried along in the wake of 
the other populations of the city, which until recently 
had only constituted the re c aya. Details should also be 
supplied regarding the first Muslim political parties 
founded in Sarajevo in this period (on these parties, 
see Popovic, op. cit., 287-9). Finally, it may be recall- 
ed that it was in Sarajevo, on 28 June 1914, that the 
Archduke of Austria Franz Ferdinand was 
assassinated by Gavrilo Princip, a young local Serb, 
and that this act served as the pretext for the 
unleashing of the First World War. 

From 1918 to 1941, Sarajevo was part of the 
Kingdom of Yugoslavia (a state which was initially 
called, and for a short time, "Kingdom of the Serbs, 
of the Croats and the Slovenians"). While continuing 
to play its role as a major regional city, and although 
it was, from 1929 onwards the centre of the Drinska 
Banovina, i.e. the Department of Drina, Sarajevo 
quite rapidly lost its former importance and was 
relegated to the second rank, suffering unfavourable 
comparisons (in all respects) in regard to the major 
centres of the country, such as Belgrade, Zagreb and 
Ljubljana. It retained its picturesque monuments and 
its pleasing aspect of an ancient Ottoman city, with its 
"upper town" composed of residential quarters and 
its "lower town", with its carsija, the streets of which 
still bore the names of the crafts which had been prac- 
tised there, its mosques, its quarters of former times, 
and its cemeteries extending over the neighbouring 
hills (in particular, one of the most spectacular of all, 
the Jewish cemetery, dating from the 16th century, 
situated on the left bank of the Miljacka). Sarajevo 
nevertheless continued to develop on the economic, 
industrial, cultural and political levels. Its population, 
within which the religious barriers were becoming 
blurred, with the consequence that an increasing 
number of mixed marriages was observed, grew from 
60,087 inhabitants in 1921 ("more than a third of 
whom are Muslims, who are for the most part crafts- 
men", F. Babinger, op. cit., in Bibl.) to slightly more 
than 80,000 in 1941. The city was naturally the prin- 
cipal religious and cultural centre not only for the 
Muslims of Bosnia-Herzegovina, but also for the 
Muslim community of the Kingdom as a whole (see 
bosna, and Popovic, 1st. balk., 312 ff. and passim). As 
regards the Muslim scholastic establishments, it 
should be noted that the medrese of GhazI Khosrew Beg 
continued to offer higher secondary education, but 
also that a new pilot scholarly establishment was open- 
ed, known as the "Academy of Islamic Law" (Serijat- 
ska gimnazija), which was the only Muslim academy in 
the Kingdom where Muslim pupils could receive an 
education comparable to that dispensed in other 
public academies. As for the Seriatska Sudacka Skola, it 
was converted in 1937 to the Visa islamska serijatsko- 
teoloska skola u Sarajevu ("Islamic High School of Law 
and Theology") and gained the status of a Faculty. 
Furthermore, the Muslim press (of various tenden- 
cies) continued to develop, as did the Muslim political 
parties (which were to disappear in 1941). (On these 
topics, see Popovic, op. cit., 328-31 and passim.) 

y the Second World War (1941-5), t] 

Pavelic, then, after the war, i 
the Federal Socialist Republic 
of Titoist Yugoslavia. In th. 
period (1945-92), Sarajevo exf 

le the capital of 

e of this latter 

:panding from some 100,000 inhabita 
1946, to 213,092 in 1961, then to about 250,000 in 
1968, and passing the figure of 300,000 in 1992, 
henceforward comprising a large number of modern 
quarters, most of them extending over the plain. The 
city continued to be the base for the guidance of the 
Yugoslav Muslim community and the seat of its chief. 
(On relations between the latter and the Communist 
authorities during this period, see Popovic, Les 
musulmans yougoslaves .) As regards the Muslim educa- 
tional establishments, they experienced several 
phases, which may be summarised thus. At the very 
beginning of the taking of power by the Communists, 
all the Muslim religious schools mentioned above 
were closed. Thus the Visa isl. ser.-teol. skola was 
definitively closed in April 1946, this coinciding with 
the abolition of the Muslim courts. Then, gradually, 
as a result of an extraordinary reversal of the situa- 
beginning with the Communist government's 



e orgam: 

predominantly Muslim — of Non-Aligned States, a 
new system was put in place. Under this system, the 
principal institution for the training of religious cadres 
became once more the renowned Gazi Husrevbeg 
medrese, then some time later, in 1977, there was estab- 
lished (still in Sarajevo) a Faculty of Islamic Theology 
(Islamski Teoloski Fakullet). The same period saw a re- 
markable flourishing of the Muslim press. 

The disintegration of Titoist Yugoslavia, following 
the collapse of the Communist world and the 
resurgence of various local nationalisms, culminated 
in the spring of 1992 in a brutal civil war in Bosnia- 
Herzegovina, in which the city of Sarajevo has 
become one of the principal theatres of operations (on 
these controversial and poorly-understood issues, as 
well as on the Muslim community, cf. X. Bougarel, 
Discours d'un ramadan de guerre civile, in L 'Autre Europe, 
26-7 [Paris 1993], 171-197; and idem, Un courant 
panislamiste en Bosnie-Herzegovine, in G. Kepel (ed.), Ex- 

,ls t 

Les apparte. 



.994, 275-99). At present, 
when, or how, the city of Sara- 
iblance of normality. 

Uography: F. Babinger, Sarajevo , in EP ; 

ija (Mula Mustafa Sevki), Ljetopis 1746- 
1804, ed. and tr. M. Mujezinovic, Sarajevo 1968; 
B. Darkot, Bosna-Saray, in tA; B. Djurdjev, Bosna, 
mEP;M. Ekmecic, Drustvo, privreda i socijalni nemiri 
u Bosni , Hercegovini (555-603), and Nacionalni pokret, 
u Bosni i Hercegovini (604-648), in Istorija srpskog 
naroda, vi/1 (1878-1918), Belgrade 1983; Evliya 
Celebi, Seyahat-name, ed. A. Cevdet, Istanbul 1315, 
v, 427-41 (see esp. E.C., Putopis. Odlonui o 
jugoslavenskim zemljama, tr. and comm. H. 
Sabanovic, Sarajevo 1967, 101-25; M. Filipovic 
and S. Corovic, Sarajevo, in Narodna Enciklopedija, 
Zagreb 1929, iv, 28-30; S. Kemura, Javne musliman- 
ske gradjevine u Sarajevu, then Sarajevske dzamije i druge 
javne zgrade turske dobe, in Glasnik Zemaljskog Muzeja, 
" ijevo 1908-11; H. Kresevljak " 


nafi , 



uprave, in Narodna Starina, vi/14, Zagreb 1927, 15- 
58; idem, Esnafi i obrti u Bosni i Hercegovini (1463- 
1878): Sarajevo ( = Zbornik za narodni zivot i obicaje 
juznih Slavena, XXX), Zagreb 1931; idem, Sarajevo 
u doba okupacije Bosne 1878, Sarajevo 1937; idem, 



Vodovodi i gradnje na vodi u starom Sarajevu, Sarajevo 
1939; idem, Esnafi i obrti u starom Sarajevu, Sarajevo 
1958; idem, Sarajevo za vrijeme austrougarske uprave 
(1878-1918), Sarajevo 1969; T. Krusevac, Sarajevo 
pod austrongarskom upravom 1878-1918, Sarajevo 
1960; M. Mujezinovic, Islamska epigrafika Bosne i 
Hercegovine, knjiga I: Sarajevo, Sarajevo 1974; Dj. Pe- 
janovic, Bibliografija stampe Bosne i Hercegovine 1850- 
1941, Sarajevo 1961; R. Pelletier, Sarajevo et sa 
region, Paris 1934; A. Popovic, L'Islam balkanique. 
Les musulmans du sud-est europe'en dans la periode post- 
ottomane, Berlin-Wiesbaden 1986; idem, Les 
musulmans yougoslaves (1945-1989). Me'diateurs el 
metaphores, Lausanne 1990; H. Sabanovic, Knjizev- 
nost muslimana Bosne i Hercegovine na orijentalnim 
jezicima, Sarajevo 1973; J. Samic, Divan de Kpimi. 
Vie et auvre d'un poete bosniaque du XVII' siecle, Paris 
1986; S. Sikiric, Sarajevske tekije, in Narodna Starina, 
vi/14, no. 1, Zagreb 1927, 77-9; V. Skaric, Srpski 
pravoslavni narod i crkva u Sarajevu u 17. i 18. vijeku, 
Sarajevo 1928; idem, Postanak Sarajevo i njegov teritori- 
jalni razvilak u 15 i 16 vijeku, in Glasnik Zemaljskog 
Muzeja, xli, Sarajevo 1929, 41-55; V. Skaric, Sara- 
jevo i njegova okolina od najstarijih vremena do austro- 
ugarske okupacije, Sarajevo 1937, Spomenica Gazi 
Husrevbegove cetiristogodisnjice, Sarajevo 1932; H. 
Tahmiscic (ed.), Poezija Sarajeva, Sarajevo 1968 (see 
esp., Sabanovic, Postanak i proslost Sarajeva pod tur- 
skom vlasc'u, 123-7); H. Tahmis&c, Sarajevo (album 
of photographs with commentary), Sarajevo n.d. 
[1969]; S. Tihic et alii, Sarajevo, in Enciklopedija 
Jugoslavije, Zagreb 1968, vii, 136-141; G. Veinstein, 
Sarajevo la Magnifique, in L'Histoire, no. 166, Paris 
(May 1993), 86-92. (A. Popovic) 

SARAKHS, a town of northern Khurasan, ly- 
ing in the steppe land to the north of the eastern end 
of the Kopet Dagh mountain chain. It was situated on 
the right or eastern bank of the Tadjant (modern 
Tedjen) river, whose uncertain flow received the 
waters of the Han Rud before finally petering out in 
the Kara Kum desert [q.v.]. According to the 
mediaeval Islamic geographers, the river bed only 
contained water at the time of floods, i.e. winter and 
early spring. Various channels were taken off the river 
for irrigation, but scantiness of water supply always 
limited agriculture there. In mediaeval times also the 
road from Nishapur and Tus to Marw passed through 

The geographers record a tale that the town was 
founded by the legendary Turkish king Afrasiyab, but 
nothing seems to be known of any pre- Islamic history. 
The first mention of Sarakhs in Islamic history is in 
22/643 when the Arab commander al-Ahnaf b. Kays 
[q.v. ] sent one of his officers to it, but this can only 
have been an exploratory probe since c Abd Allah b. 
c Amir [q.v.] in 31/651-2 led a campaign into 
Khurasan, capturing Nishapur and other towns as far 
as Sarakhs (al-Tabarl, i, 2682, 2884, 2887-8). Sarakhs 
is mentioned during the fighting in Khurasan be- 
tween Abu Muslim's partisans and the last Umayyad 
governor there, Nasr b. Sayyar [q.v.], and in 185/801 
there took place at Sarakhs and at neighbouring Nasa 
[q.v.] a rebellion against the oppressive c Abbasid 
governor C A1I b. c Isa b. Mahan [q.v.] led by the mawld 
Abu 'l-Kiiasib Wuhayb b. c Abd Allah (183-6/799- 
802) (see E.L. Daniel, The political and social history of 
Khurasan under Abbasid rule 747-820, Minneapolis and 
Chicago 1979, 171). The geographers of the 4th/10th 
century describe Sarakhs as a considerable town, half 
the size of Marw, with a Friday mosque, good 
agriculture, including grain grown for export to 
Nishapur, and extensive pasture grounds for camels 

and sheep. Within the population, so al-Mukaddasi 
states, there were two factions of the Hanafi 'Arusiyya 
and the §hafi c T Ahliyya (see C.E. Bosworth, The Ghaz- 
navids, 165-6). 

Sarakhs played a significant role in the overrunning 
of Khurasan by the Turkmens in the first half of the 
5th/llth century. In 425/1025 Mahmfld of Ghazna 
allowed 4,000 Oghuz families and their herds to cross 
the Oxus and settle near Farawa, Abiward [q.vv.] and 
Sarakhs, but by 428/1036 the Oghuz were demanding 
a grant of the revenues of Marw, Abiward and 
Sarakhs. It suffered badly from the devastations of the 
Turkmens, so that when Mas'ud of Qhazna appeared 
there with his army in 431/1040, the exasperated in- 
habitants refused him entry, and Mas c ud had to storm 
the citadel, killing many of the people; it was thus 
from Sarakhs that the sultan set forth for his ill-fated 
battle with the Saldjuks and their forces at Dandankan 
[q.v. in Suppl.] in the waterless desert between 
Sarakhs and Marw (see Bosworth, The Ghaznavids, 
224, 250-1, 265). In the second half of the 5th/12th 
century and after the end of Sandjar's sultanate, 
Sarakhs was held by the Oghuz chief Malik Dinar, 
and then by the Kh w arazmian claimant to power, 
Sultan Shah b. II Arslan. During the time of the 
Mongol invasions, Cingiz Khan in 618/1221 sent his 
son Toluy to occupy the towns of Khurasan, including 
Sarakhs: it submitted and received a Mongol shihna, 
but rebelled, like other towns in the province, on hear- 
ing rumours of the Kh w arazm-Shah Djalal al-Dln's 
successes (Djuwaynl-Boyle, i, 155-6, 162, 301). The 
town must nevertheless have slowly revived after the 
Mongol devastations. Ibn Battiita [q.v.] passed 
through it without mentioning anything except 
Sarakhs' s connection with the Sufi shaykh Lukman al- 
Sarakhsl (whose gunbad or tomb still exists in the town) 
(Rihla, iii, 79, tr. Gibb, iii, 583), but Hamd Allah 
Mustawfi (mid-8th/14th century) describes the town 
as having a strong wall 5,000 paces in circumference 
and a flourishing agriculture, especially of melons and 
grapes (Nuzha, ed. Le Strange, 159, tr. 155). 

During the period of Safawid-Ozbeg warfare, it had 
an exposed position in the frontier zone between the 
two rival powers. In 932/1526 c Ubayd Allah Khan 
Shrbani occupied Sarakhs en route for his campaign 
against Mashhad and Tus (Babur-nama, tr. Beveridge, 
534). The raids of the Tekke Turkmens of Marw on 
Persian territory did not cease until after the Persian 
government in ca. 1850 constructed a strong fort at 
Sarakhs, on the left or western bank of the Tedjen 
river, shortly after which a new threat appeared when 
the Russians moved into Central Asia and built a 
military post and settlement at Old Sarakhs on the 
right bank (G.N. Curzon, Persia and the Persian ques- 
tion, London 1892, i, 195-8). 

Modern Persian Sarakhs (lat. 36° 32 'N., long. 61° 
07' E.) is the chef-lieu of a ba khsh of the same name 
in the shahrastan of Mashhad in the province (ustan) of 
Khurasan; in ca. 1950 it had a population of 5,000 
(Razmara, Farhang-i djughrdfiya-yi Iran-zamin, ix, 212- 
13), which' had increased by 1991 to 22,247 
(Preliminary results of the 1991 census, Statistical Centre 
of Iran, Population Division). 

Bibliography (in addition to references given in 
the article): A. Burnes, Travels into Bokhara. . . , Lon- 
don 1834, ii, 50-3; C.E. Yate, Khurasan and Sistan, 
Edinburgh and London 1900, 34; Le Strange, The 
lands of the eastern Caliphate, 395-6; c Abbas Sa c IdI, 
Sarakhs diruz wa imruz, pazhuhishi dar djughrdfiya-yi 
tarikhiwa ansdbi-yi Sarakhs, Tehran 1354/1975; W. 
Barthold, An historical geography of Iran, Princeton 
1984, 62-3. (C.E. Bosworth) 



Tayyib b. Marwan, th( 
of al-Kindl and, like 
vocate of Greek 1 
its Muslim integi 



d adab. 

with a 

i 220/835 and died early in 
286/899. A fact attested for his obscure early life is his 
participation as the delegate of al-Kindl in a 
multireligious philosophical-theological debate about 
Christianity and the Trinity (see Moosa and 
Holmberg). He began his career as an educator of the 
future caliph al-Mu c tadid. In 271-2/885, he accom- 
panied the prince on an unsuccessful military expedi- 
tion to al-Tawahin in southern Palestine, keeping a 
journal detailing its itinerary. He became an influen- 
tial nadim \q.v.} of the caliph and, in 282/895, was ap- 
pointed to the hisba [<?.».] and other offices. This, how- 
ever, may have contributed to his downfall. Already 
in the following year, he was incarcerated. He died, 
or rather was put to death, in prison three years later. 
The reason for this turn in his fortunes was a mystery 
that was much debated at the time and continued to 
intrigue scholars through the centuries. Many ex- 
planations were put forward, none of them provable. 
The unwise betrayal of a high-level secret, a falling- 
out with the caliph himself, and also rumours of 
"heresy", might have contributed. He appears in- 
deed to have unwisely expressed objectionable views, 
including seeming doubts about the credibility of pro- 
phets (see also below). His supposed influence on the 
caliph in religious matters might have become an em- 
barrassment to the latter, once he was given highly 
visible official positions. And there were no doubt the 
usual rivalries at court; he himself is described as 
fiercely protective of his position (see al-Safadi, Wdfi, 
xii, 17, under al-Hasan b. Abi '1-Ra c d al-Khurasanl). 
His great productivity as a scholar and writer may 
in part have been as transmitter of al-Kindl, albeit, it 
seems, a rather creative one; this is particularly likely 
where identical titles on philosophical subjects are at- 
tested for both only in late bibliographical tradition. 
No copy of a complete work of his has been authen- 
ticated so far. The recent discovery of a Kitab Addb al- 
muluk also remains doubtful for the time being, even 
though the title appears among his works; the attribu- 

scripts now known, and the work itself is an in- 
teresting reworking of the Akhlak al-muluk by 
Muhammad b. al-Harith al-TaghlibI/Tha c labi, pub- 
lished earlier as Kitab al-Tadf. with a wrong attribution 
to al-Djahiz (see G. Schoeler, in ZDMG, cxxx [1980], 
217-25). On the other hand, we have a cornucopia of 
quotations from his works or attributed to him as a 
transmitter; a noteworthy recently published source is 
Aba Hayyan al-Tawhldi, BasaHr, ed. Wadad al-Kadl, 
Beirut 1408/1988. Geography, in particular, was 
enriched by his above-mentioned journal; it was 
discovered in al-Mu c tadid's library under interesting 
circumstances and acquaints us with a unique docu- 
ment of early Muslims geography. He is credited with 
other geographical and topographical writings. He 
and al-Kindt are said to have provided the materials 
for a world map (see M. Kropp, in Proceedings of the 
Ninth Congress of the Union Europe'enne des Arabisants et 
Islamisants, Leiden 1981, between 160 and 161). He 
wrote on the full extent of the paideia of the nadim, on 
cooking and politics, on love and music, among many 
other subjects. Fragments on the theoretical aspects of 
music are preserved in al-Hasan b. Ahmad al-Katib, 
Kamdl adab al-ghina>, ed. Cairo 1975, v. musiki, vii, 
683-4; Fr. tr. Amnon Shiloah, La perfection des con- 

ks, Paris 1972). He occurs frequently 
of anecdotes on singers and wits. His 
interest in comparative religion is further attested by 
his report on the Sabians on the basis of al-Kindl pre- 
served in the Fihrist (another fragment in c Abd al- 
Djabbar, Mughni, v, 152). His hilarious spoof that 
targeted the anti-Greek religious bias of a narrow- 
minded member of the Ibn Thawaba family [q. v. ] as 
told by Abu Hayyan al-Tawhidi, Akhlak al-wazirayn, 
ed. Ibn Tawit al-Tandji, Damascus 1385/1965, 235- 
47, may possibly have been concocted or embellished 
by al-Tawhidi himself; be this as it may, its ascription 
to him as well as his writing on Sabianism could easily 
have added to impeaching his orthodoxy for later 
generations. All his works and ideas (among them the 
invention of a transliteration system for foreign 
languages, see Hamza al-Isfaham, Tanbih, ed. M.H. 
Al Yasin, Baghdad 1968, 35) reveal a lively thinker 
and (to a degree) free spirit who probably had few 
equals in his time. He may well be considered as 
representative of intellectual currents in contem- 
porary Baghdad that were about tc 
Bibliography: The n 

orks a 

: the Fihr 


to change direction. 

trees for his life and 

condarily, Ibn Abi 

is c udl; KiftT; 

Usaybi'a. Cf. further, for i 

Yakut, Irshdd, i, 158-60; DhahabI, Siyar, Beirut 
1412/1992, xiii, 448-9; Safadi, Waft, vii, 5-8; Ibn 
Hadjar, Lisdn, i, 198-9; Brockelmann, F, 231-2, 
S I, 375; Sezgin, GAS, iii, 259, v, 263, vi, 162-3, 
vii, 137, ix, 233. Numerous other biographical 
references are almost exclusively concerned with the 
circumstances of his death. See F. Rosenthal, 
Ahmad b. at-Tayyib as-Sarahsi, American Oriental 
Series 26, New Haven 1943; idem, in JAOS, lxxi 
(1951), 135-42, lxxvi (1956), 27-32, lxxxi {1961), 
222-4; idem, articles to be published on Addb al- 
muluk and on the Rangstreit between lovers of boys 
and of girls; Matti Moosa, in al-Madjalla al- 
Batriyarkiyya (The Patriarchal Magazine, Organ of the 
Syrian Orthodox Patriarchate of Antioch and All the East), 
vii (1969), 189-97, 244-52; idem, in JAOS, xcii 
(1972), 19-24; B. Holmberg, A treatise on the unity and 
Trinity of God by Israel of Kashkar, Lund Studies in 
African and -Asian Religions 3, Lund 1988, 50 ff., 
84 ff.; Mudjtaba Minuwl, in L^awldan Khirad, i 
(1354/1975), 9-18 (no more publ.; reference provid- 
ed by Said Arjomand). (F. Rosenthal) 
al-SARAKHSI, Muhammad b. Ahmad b. Abi 
Sahl Abu Bakr, Shams al-A>imma, a Hanafi jurist 
of the 5th/llth century, who lived and worked in 
Transoxania, inheriting and developing the juristic 
tradition of that region. He produced a number of 
works, the most important being the Mabsut, the Shark 
al-Siyar al-kabtr, and the UsUl al-fikh. The first of these 
is a work of furu\ a commentary on the mukhlasar 
of Muhammad b. Muhammad al-MarwazT (d. 
334/945). This in turn was an epitome of the works of 
Muhammad b. al-Hasan al-Shaybam, the founda- 
tional texts of the Hanafi tradition. Al-SarakhsT re- 
introduced and explored the rules of al-Shaybam, 
organising his material around points of dispute (ikh- 
tilaf) and incorporating information related to local 
Hanafi tradition and other schools of law, apparently 
derived from oral transmission and local teaching 
practice. His organisation, comprehensive coverage, 
exploration of ikhtildf, and manipulation of 
hermeneutical argument, all conduce to make this 
work a remarkable achievement of juristic literature. 
It remained a point of reference for the developing 
Hanafi >ru c tradition till the 19th century. The Shark 
is a commentary on the Kitab al-Siyar al-kabtr of al- 
Shaybam. In its published form it is not always easy 


to distinguish the text and the commentary (also true 
of the Mabsut), but it too demonstrates an overall con- 
cern for comprehensive coverage, development of 
rules and considered hermeneutical argument. Al- 
Sarakhsl's Usui draws on the independent Transoxa- 
nian Hanafi tradition represented by Abu '1-Hasan al- 
Karkhl (d. 340/951) and Ahmad b. Muhammad al- 
Shashi (d. 344/955) and on the usul works of al- 
Djassas (d. 370/980). He also refers to the Risdla of al- 
Shafi c i, the opinions and some writings of later Shafi'T 
thinkers (e.g. Ibn al-Suraydj, d. 306/918), and to the 
major stances of other traditions. 

The Hanafi biographical tradition has little in- 
dependent information on al-Sarakhsi's life. It draws 
on and elaborates clues supplied in the works. The In- 
troduction to the Mabsut and the concluding formulae 
of a number of its sections reveal that the author dic- 
tated it from prison. The Sharh identifies c Abd al- c Aziz 
b. Ahmad al-Hulwanl as a teacher, and the Usui 
begins with the statement that it was dictated in Uzdj- 
and or Ozgend in 479/1086. Biographical notices 
emerge later e.g. in the Kitdb al-Djawahir al-mudiyya of 
c Abd al-Kadir b. Abi 'l-WafS' Muhammad b. 
Muhammad al-Kurashl (d. 775/1373) and the Tddjal- 
tarddjim of al-Kasim Ibn Kudubugha (d. 879/1474). 
In addition to conventional and formulaic items, al- 
Kurashi states that al-Sarakhsi died ca. 490/1096, and 
names three students, Muhammad b. Ibrahim al- 
Hasiri, c Uthman b. c Ali al-Baykandl and c Umar b. 
Habrb, grandfather of al-Marghinani, the author of 
the Hiddya. Ibn Kutlubugha expands the story of the 
imprisonment, adds that al-Sarakhsi ended his life in 
Farghana under the protection of the Amir Hasan, 
suggests a date of death about 500/1106, and gives an 
anecdote about a local amir and his umm walads. The 
anecdote is intended to demonstrate the knowledge, 
skills and integrity of al-Sarakhsi, but Heffening, in 
EP , discovered in it a cause for his imprisonment. 
Abu '1-Hasanat Muhammad c Abd al-Hayy al- 
Lakhnawi, writing in 1293/1876, summarised the 
biographical tradition, incorporating a few more 
details (e.g. he identifies another student, Muham- 
mad b. c Abd al- c Az!z al-Uzdjandi, grandfather of 
Kadikhan) and citing an aberrant death date of 
438/1046. HadjdjI Khalifa, v, 363, gives 483/1090 as 
date of death and has been followed by Brockelmann, 
Heffening and Sezgin.- 

Bibliography: Sarakhsi, K. al-Mabsut, 30 vols., 
Cairo 1324 (repr. Beirut, n.d.); Sharh al-Siyar al- 
kabir, 5 vols., ed. Salah al-DIn al-Munadjdjid, 
Cairo 1951 (repr. 1971); Usui al-fikh, 2 vols., ed. 
Abu '1-Wafa 3 al-Afghanl, Haydar'abad 1372; 
Kurashi, al-Djawahir al-mudiyya, 2 vols., 
Haydarabad 1332; Ibn Kutlubugha, Tddj al- 
tarddjim, ed. FlQgel, Leipzig 1862; Lakhnawl, al- 
FawaHd al-bahiyya, PHaydarabad 1324 (repr. Beirut 
n.d.); Heffening, EP art., s.v., Brockelmann, 
GAL, I, 460, Suppl. 1, 638; Sezgin, GAS, i, 423, 
424, 428, 430 etc. (N. Calder) 

SARAKUSTA, Saragossa, a town situated on 
the river Ebro in Spain, regional capital of its 
eponymous province and of the current Communidad 
Autonoma de Aragon. Founded in 24 B.C. by Augustus 
as a Roman military colony, on the Iberian site of 
Salduba, it was called in Latin Caesarea Augusta, a 
name corrupted into the form Cesaragosta, which was 
adopted, virtually unchanged, by the Muslims after 
their conquest of the Iberian Peninsula; it is transcrib- 
ed into Arabic as Sarakusta (nisba: Sarakusti). The cor- 
respondences s > z and k > g and the current 
assimilation -st- > -z- explain its modern form in 
Spanish, Zaragoza. 

Sarakusta was one of the most important cities 
(madina) of al-Andalus, between 95/714 and its con- 
quest by Alfonso I of Aragon (512/1118), considered 
the regional capital (hddira) or the "metropolis" of the 
"Upper March" of al-Andalus (umm al-thaghr al-aHa), 
the ihaghr or northern frontier extending in principle 
to the north of the Pyrenees, and, after the Christian 
conquests as far as Pamplona (captured in 183/799) 
and Barcelona (captured in 185/801). This was fixed, 
for the duration of the three remaining centuries, in 
the valley of the Ebro, with the zones (iklim) of 
Sarakusta and of Tudela (Tutila), Huesca (Washka), 
Barbastro (Barbitaniya), Lerida (Larida) and 
Calatayud (Kal c at Ayyub), in addition to the eastern 
zone of Tortosa (Turtusha) and the southern zone of 
Barusha, bordering on the "Middle March" (al-thaghr 

The zone (iklim) of Sarakusta included the districts 
(ndhiya) of the city itself (al-madina), and others such as 
Belchita (Balshad), Cazarabet (Kasr c Abbad), Cutan- 
da (Kutanda), Fuentes (Funtish), Gallego (Djallik), 
Jalon (Shalun), Pleitas (Baltash), and Zaydun. It 
comprised fortresses (hisn), villages (balda) and 
hamlets (karya), denominations applied not always 
systematically to places in this zone, such as Alcaniz 
(Kannish), Almenara (al-Manara), Caspe (Kashb), 
Calanda (Kalanna), Montanana (Munt Anyat), Ricla 
(Rikla), Rueda de Jalon (Ruta), and Zuera 

The town plan of Sarakusta included a space 
enclosed within a wall of stone, built by the Romans 
and preserved throughout the four centuries of 
Muslim domination; this wall surrounded an ir- 
regular rectangle of approximately 600x900 m 
where, according to the calculations of L. Torres 
Balbas, some 17,000 inhabitants lived. This intra- 
mural space was crossed by two perpendicular 
highways (the former cardo and decumanus) which con- 
nected four gates: the Gate of the Bridge (Bab al- 
Kaniard), on the river Ebro; to the east, the Gate of al- 
Kibla, or of Valencia; to the west the Gate of Toledo, 
or "of the Jews" (Bab al-Yahud), and to the south the 
Gate of the Sinhadja, indicating a settlement of these 
Berbers, a name still evident in the "Cinegio" Arch. 
In the north-west corner of the enclosed space there 
was a fortified compound, the seat of authority (al- 
Kasr), known as al-Sudda (like the Sudda of Cordova), 
a name currently born by the "Torreon de la Zuda". 
Also within the enclosed space stood the Great 
Mosque (al-Dj_dmi°), which according to an im- 
probable tradition is said to have been founded by a 
number of venerable individuals (sahdba or tdbi'un) 
who were supposed to have arrived in Sarakusta with 
the vanguard of the Muslim conquerors: archaeology 
has proved that this Great Mosque was built over a 
Roman temple, dating from the time of Tiberius, 
which became a church with the arrival of Christian- 
ity. Some Andalusian sources indicate that the Great 
Mosque of Sarakusta was enlarged twice, in 242/856- 
7 and ca. 409-12/1018-21. Recent archeological ex- 
cavations have revealed the rectangular shape (54 x 86 
m) of this Mosque, with a pillar-supported hall of nine 
naves, comparable in dimensions to the greatest mos- 
ques of al-Andalus, those of Cordova and Seville. All 
that remains of it are the lower portions of the 
minaret, and a few capitals dating from the 5th/ 11th 
century and uncovered in excavations carried but in 
what is now the Seo del Salvador, the church con- 
structed by the Christians on the site of the Great 
Mosque of Sarakusta. 

Outside the wall there were extensive suburbs 
(rabad), such as that known as "the Tannery" (al- 


Dibagha, currently the suburb of "Altabas", formerly 
"Atabahas", on the other side of the river), several 
named after the Sinhadja (outside the eponymous 
Gate), and other suburbs which encompassed the wall 
in its entirety, in their turn defended by a wall of clay 
(radam), with its own small gates, five of which are 
known to us by later names belonging to the Christian 
period. Outside the Gate of Sancho, to the north- 
west, stretched the great public esplanade of the 
Musara or Musalla (a name retained in what is now 



Outside this wall of clay is located the most impor- 
tant building of Sarakusta which is still standing: the 
castle-palace of Aljaferia (al-DjaHariyya). named after 
Abu Dja'far, kunya of the king of the f&Hfa of 
Sarakusta, al-Muktadir (second half of the 5th/l 1th 
century), who built the palace alongside a tower 
dating from the Umayyad period (currently "Torre 
del Trovador") which possibly has Roman founda- 
tions. The builder-king celebrated his palace in his 
poetry, calling it "house of joy" (dar al-surur); his 
honorific title of al-Muktadir is to be found inscribed 
on a capital of the Aljaferia. preserved in the Museum 

Sarakusta, known as the "White City" (al-madlna 
al-bayda*), appeared, according to the Arab 
geographers, as "a white stain surrounded by the vast 
emerald green of its countryside". In fact, the sources 
praise the quality and the abundance of its 
agriculture, well irrigated by the great river Ebro and 
its numerous tributaries. Commerce found in 
Sarakusta "the gateway to all routes", according to a 
reference in the sources. There were salt mines and a 
thriving fur trade, producing some renowned furs 
known as sarakustiyya. 

In this space, the Muslim life of Sarakusta 
developed during the four centuries (2nd-8th to 6th- 
12th) of Islamic political domination, also during the 
following four centuries (from 1118 to 1614) of the 
well-documented presence of "Mudejar" [q.v. ] and 
"Moriscos" [q.v.], the "Moors", subjected to Christ- 
ian political power. The four centuries of Muslim 
political domination reflect the general chronology of 
the history of al-Andalus [q.v.]: (a) conquest and rule 
by amirs dependent on the Orient and Ifrikiya; (b) the 
independent Umayyad dynasty of al-Andalus; (c) the 
kingdoms of to/as (muluk al-tawaHf); and (d) the 
Almoravids. Nothing certain is known regarding the 
Muslim conquest of Sarakusta, possibly in 95/714; 
Christian sources, such as the Cronica Mozdrabe, speak 
of fierce resistance on the part of the town, but two 
Christian churches remained in use there throughout 
the Islamic period: the church of Santa Maria (with 
the pious tradition of the apparition of the "Pillar") 
and the church of Las Santas Masas, centres of an ac- 
tive Christian presence in the town. There was also a 
Jewish community, to which sporadic reference is 

Sarakusta takes a prominent role in the history of 
al-Andalus from the year 124/742 onwards, becoming 
embroiled in the struggles of the baladiyyun Arabs 
against the Shamiyyun, and of the "North Arabs" 
against the "South Arabs", the latter forming the ma- 
jority in the Upper March. In 132/749-50, the gover- 
nor of al-Andalus Yusuf al-Fihri [q.v.] sent to 
Sarakusta as wall a certain al-Sumayl, a "North 
Arab", in the hope of exercising better control over 
the "South Arab" majority; the latter rose in revolt 
four years later, and ultimately supported the can- 
didature of the Umayyad c Abd al-Rahman I [q.v.], 
but after the latter had become amir of al-Andalus, 
these "Southern Arabs", dominating the Upper 

March, rebelled incessantly against the central power 
of al-Andalus, even seeking the aid of Charlemagne, 
who came to Sarakusta in 778 but found the gates of 
the city closed against him; whilst withdrawing, the 
Frankish army was attacked, this constituting some of 
the most renowned episodes recorded in the Chanson de 

These frontier regions of al-Andalus persisted in 
their autonomist tendencies, rejecting both 
dependence on the Christians and dependence on 
Cordova, which sent its armies there on numerous oc- 
casions, commanded in person by successive amirs of 
al-Andalus, who succeeded, periodically, in controll- 
ing the region by installing Cordovan governors, an 
unsatisfactory measure in terms of internal ad- 
ministration, and especially in terms of the external 
Christian threat. Collaboration with local families, in 
principle loyal to the Umayyads, was the major 
recourse of the central power, from the later years of 
the 2nd/8th century onwards. The muladi (muwallad 
[q. v. ]) family of the Band Kasi [q. v. ] were to become 
periodically the masters of Sarakusta, leading pro- 
tagonists, probably, in the activities of the Upper 
March, alternating loyalty and rebellion towards the 
layyad amirs, until the decline of muladi power 


t al-Andalus, and il 

ment by a new power exercised by certain Arab 
families, who were relied upon by the Umayyads from 
the later years of the 3rd/9th century onwards, as hap- 
pened as Sarakusta with the Banu Tudjib or 
Tudjlbiyyun [q.v.], in power between 276/890 and 
430/1038, such families alternated between sub- 
missiveness and an autonomism which was total after 
the civil war (fitna) of the early 5th/llth century, 
leading to the establishment of the first dynasty of the 
laifa of Sarakusta, which was replaced by the dynasty 
of the Banu Hud [q.v.], from 430/1038 until the con- 
quest by the Almoravids in 503/1 110. The latter were, 
however, unable to maintain Muslim political 
domination at Sarakusta for more than eight years, 
until 512/1118. The Muslim inhabitants remained 
there after the Christian conquest, subject to the 
regulations imposed upon "Mudejars" [q.v.] and 
"Moors" [q.v.], until their expulsion in 1614. 

Cultural life came into being very early at 
Sarakusta, given its status as an important urban cen- 
tre, with the first manifestations in the 3rd/9th cen- 
tury, and with consolidation in the following century, 
centred on cultured families such as the Banu Furtish 
and the Banu Thabit, among others. During these 

cultural isolation from the rest of al-Andalus but 
maintained direct relations with the Orient and the 
Maghrib, especially through the journey of the 
Pilgrimage. The cultural flowering of the 5th/ 11th 
century was, in part, the result of the arrival of major 
poets and scribes from Codova, making their way to 
the great laifa of the Upper March, under the 
patronage of the rulers of the Banu Tudjib and Banu 
Hud. In the 6th/12th century, the finest flowering of 
the culture of Sarakusta was to be seen amongst its 
citizens in exile, as in the case of Ibn Badja [q.v.], but 
in the city and its neighbourhood the culture of the 
Mudejars and the Moors also survived, manifesting 
itself in aljamia (al- c adj_amiyya). 

Bibliography: 1. Sources. RazI, tr. E. Levi- 
Provencal, in Al-And. (1953), 78; Ibn Hawkal, 
Leiden 1938-9, 115-16; IdrisI, Description, ed.-tr. 
Dozy, 190-1; Uns, ed.-tr. J. A. Mizal, Madrid 
1989, n. 209; c Udhri, Madrid 1965, 21-5; tr. F. de 
la Granja, La Marca Superior, Saragossa 1967, 11-17; 
HimyarT, al-Rawd al-miHar, ed.-tr. Levi-Provencal, 


317/1118; Yakut, Mu'djam, s.v.; Crdnica mozdrabe de 
754, ed.-tr. Saragossa 1980, 70-2; Ibn Hayyan, 
Muktabis, ii, Beirut 1973, 697; iii, Paris 1937, 1165; 
v, Madrid 1979, tr. Madrid-Saragossa 1981, 446; 
vii, tr. Madrid 1967, Beirut 1965, 306; Ibn Hazm, 
Djamhara, tr. E. Teres, in Al-And. (1957), 376; Ibn 
Bassam, al-Dhakhira, Beirut 1979, ii, 998, iv, 890, 
vi, 942, vii, 423; Ibn al-Abbar, al-Hulla, Cairo 
1961, ii, 417; Ibn Sa c Id, al-Mughrib, Cairo, ii, 557- 
8; Ibn al-Athir, al-Kdmil, tr. E. Fagnan, Annates, in- 
dex; Dhikr, ed.-tr. L. Molina, Madrid 1983, ii, 297; 
Marrakushl, al-Mu%ib, ed. Dozy, 41, 50, 85, 148; 
Ibn 'Idhari, al-Baydn, Paris 1930 and Beirut 1967, 
passim; Ibn Khaldun, TaHikh, Beirut 1983, viii, 
659-60; al-Djazzar, Diwdn, ed. S. Barbera, thesis, 
Universite Complutense, Madrid 1991, unpubl.; 
Makkarl, Nafh, Beirut 1968, viii, index. 

2. Studies. M.J. Viguera, Aragon musulmdn, 
Saragossa, 2 1988 (exhaustive bibl. at 249-86); C. 
Laliena and P. Senac, Musulmans et Chretiens dans le 
Haul Moyen Age; aux origines de la Reconquete aragonaise, 
Montrouge 1991; E. Manzano, La frontera de al- 
Andalus en epoca de los Omeyas, Madrid 1991; idem, 
La Marche Superieure d'al-Andalus et I'Occident chretien, 
Madrid 1991; A. Turk, Ibn c Ammdr: unafigura tipica 
del siglo XI, in Revista de Historia Jeronimo Zurita, lxiii- 
lxiv (1991), 141-69; Simposio Internacional sobre la 
Ciudad Isldmica, Saragossa 1991; B. Cabanero, Los 
restos isldmicos de Malejdn (Zaragoza), Saragossa 1992; 
J. A. Souto, Textos drabes relativos a la mezquita aljama 
de Zaragoza, in Madrider Mitteilungen, xxx (1989), 
391-426; idem, El problamiento del tirmino de Zaragoza 
(siglos VIII-X), in Anaquel de Estudios Arabes, iii 
(1992), 113-52; idem, Restos arquitectonicos de epoca 
isldmica en la Seo del Salvador (Zaragoza), in Madrider 
Mitteilungen, xiv (1993), 308-24; Los cementerios de 
Zaragoza, Saragossa 1991 ; Atlas de Historia de Aragon, 
Saragossa 1992; P. Senac (ed.), Frontieres el espaces 
pyreneens au Moyen Age, Perpignan 1992, 129-50; 
Boletin Aljamia, vi (1994), 10, 19, 21, 23, 30. 

(M.j. V.GUERA) 
al-SARAKUSTI, the nisba of t • - - ■ 

, father 

i, both 

the northern Spanish town of Sarakusta [q.v.] or 
Saragossa. These are Abu Muhammad Kasim b. 
Thabit b. Hazm b. c Abd al-Rahman b. Mujarrif b. 
Sulayman b. Yahya al- c Awff al-Zuhri (255-302/869- 
914) and his father Abu '1-Kasim Thabit (217- 
313/832-925 or 314/926). The biographical sources 
mention variants in their nasab that show that their 
genealogy was manipulated. They were Berbers who 
had established ties of wald? (wald? '■aldka) with the 
Banu Zuhra, as all the Berbers of the Upper Frontier 
in al-Andalus had done. Kasim's father Thabit, angry 
for some unknown cause, decided not to use the nisba 
al-Zuhri and adopted instead the nisba al- c Awfi. This 
way he pretended that his ancestors were clients of a 
descendant of the Prophet's Companion c Abd al- 
Rahman b. c Awf al-Zuhri [q.v.], who was allegedly in 
the Iberian Peninsula at the time of the conquest. 

Kasim and his father are credited with having intro- 
duced into al-Andalus al-Khaltl's Kitab al- c Ayn, but 
they are especially famous for their work on gharib al- 
hadith (before them, the Andalusians c Abd al-Malik b. 
Hablb and Ibn c Abd al-Salam al-Khushanl had 
already written books on the same genre). Both 
travelled to the East in the year 288/900, studying in 
Misr (Old Cairo) and Mecca with, among others, the 
traditionists al-Nasa'i, al-Bazzar and Ibn al-Djarud (a 
complete list of their teachers is given by Ibn Harith 
al-Khusham; see also al-Fahham in RAAD, li [1976], 
512-3). In 294/906 they returned to al-Andalus as ex- 

perts in hadith and lugha. Kasim commenced work on 
his al-DaldHl fi shark ma aghfala Abu ^Ubayd wa-bn 
Kutayba min gharib al-hadith, but died before he finished 
it. The book was completed by his father Thabit. Al- 
Zubaydl reports that some said they were not the real 
authors but rather it was the work of an Eastern 
author. In any case, it was much lauded, all the An- 
dalusian biographers repeating Abu C A1T al-Kali's 
praise, stressing that it was unparalleled even in the 
East (we are here at the beginning of the development 
of an Andalusian sense of local pride well reflected in 
Ibn Hazm's Risdla fi JaddHl al-Andalus where of course 
the K. al-DaldHl has a place of honour). Organised ac- 
cording to the musnad system, the K. al-DaldHl is par- 
tially preserved in three mss.: BG Rabat, no. 197 kdf 
(microfilm in the Ma c had al-makhtutat); Zahiriyya, 
no. 1579 = no. 41 lugha (microfilm in Maktabat al- 
Awkaf, Baghdad); Turk Evkaf Muzesi, Istanbul, no. 
1682. They have been carefully described by al- 
Fahham, who is also responsible for a partial edition 
of the text. Muhammad b. Aflah, mawld of al-Hakam 
II, wrote a TaHik <ala '1-DaldHl (see M. al-Manunl, 
Thakajat al-sakdliba hi 'l-Andalus, in Awrdk, v-vi [1982- 
3], 27, no. 33). The well-documented transmission of 
al-DaldHl through different chains reaches up to the 
AndalusI Ibn Salim al-Kala c I (7th/13th century). 

Kasim is described as ascetic and pious. Pressed by 
his father, who had been a judge in Saragossa, to ac- 
cept the same post, Kasim asked for three days to con- 
sider the matter and died in the meantime. On the 
basis of this anecdote, the historical value of which is 
open to discussion, he is credited with being a mugjdb 
al-da' i wa, for he would have succesfully asked God to 
spare him the dangerous office of judge. 

There is information about other members of the 
family who specialised in the transmission of the K. al- 
DaldHl, for example Kasim's son Thabit (289- 
352/902-63), who wrote a copy of the work for al- 
Hakam II. The last known descendant is Thabit b. 
c Abd Allah b. Thabit b. SaTd b. Thabit b. Kasim b. 
Thabit (d. 514/1120). 

Bibliography: For the primary sources, see M. 
Fierro, Historia de los autores y transmisores andalusies 
(HATA), forthcoming. See also Kahhala, iii, 99- 
100, viii, 96-7; Zirikli, ii, 97, v, 174; J. Vernet, El 
valle del Ebro como nexo entre Oriente y Occidente, in 
Boletin de la Real Academia de Buenas Letras de 
Barcelona, xxiii (1950), 249-86, nos. 181, 200, 201 
(repr. in J. Vernet, De <Abd al-Rahmdn I a Isabel II, 
Barcelona 1989); M.A. Makki, Ensayo sobre las apor- 
taciones orientates en la Espana musulmana, Madrid 
1968, 156, 198-9, 264; M" J. Hermosilla, Las obras 
andalusies en la Gunya del Qadi Hyad, in Anuario de 
Filologia, v (1979), 192, no. 2; GAS, viii, 252; L. 
Molina and M a L. Avila, Sociedad y cullura en la 
Marca Superior, in Historia de Aragon, iii, Saragossa 
1985, 90; M. Marin, Nomina de sabios de al-Andalus, 
in EOBA, i (Madrid 1988), nos. 335, 1051; L. 
Molina, Familias andalusies, in ibid., ii (Granada 
1989), 69-70; J.M.F. Vizcaino, Familias andalusies 
en la Fahrasa de Ibnjayr, in ibid., v (Madrid 1992), 
473, no. 5. On the K. al-DaldHl, see c Izz al-Dln al- 
Tanukhi, KSsim b. Thabit al-Sarakusti wa-kitdbuhu fi 
gharib al-hadith al-musammd hi 'l-Dala'il, in RAAD, 
xli (1966), 3-20; Sh. al-Fahham, Kitab al-DaldHl fi 
gharib al-hadith li-Abi Muhammad Kasim b. Thabit 
al-<Awff al-Sarakusti, in ibid. , 1 (1975), 75-1 10, 303- 
21, 512-27, li (1976), 232-94, 481-517; idem, Hadith 
al-Sha c bi fi sifat al-ghayth. Nass mustakhradi min Kitab 
al-Dala'il fi gharib al-hadith li-Abi Muhammad Kasim 
b. Thabit al-'-Awfi al-Sarakusti, in ibid., lviii/1 (1983), 
3-69. (Maribel Fierro) 

SARANDIB, the name given in mediaeval Islamic 
geographical and historical sources to the island of 
Ceylon (modern Sri Lanka). The Arabic form 
renders well the Skr. Simhala "Ceylon" + dvipa 
"island"; an intermediate form is found in al-Biruni, 
India, tr. E. Sachau, London 1910, i, 233, as 
Sangaladtp. By the time of Yakut (early 7th/13th cen- 
tury), the form Silan is found (Buldan, ed. Beirut, i, 
346, art. Bahr al-Hind). 

Most of the mediaeval Islamic geographers, from 
Ibn Khurradadhbih onwards, give some account of 
Sarandlb, placing it in the Sea of Harkand ( = the Bay 
of Bengal) between India and China and describing it 
as the last of the DTbadjat (an Arabised plural form of 
Skr. drnpa), i.e. the Indian Ocean archipelagos of the 
Laccadives, Maldives [q. tin.], etc. An authority like al- 
Idrisi gives an exaggeratedly large size for Sarandlb, 
an over-estimation going back to Ptolemy and noted 
by Marco Polo in his section on Seilan (Yule-Cordier, 
The Book of Ser Marco Polo, London 1903, ii, 312-30, 
chs. xiv-xv). For the Arab and Persian writers, the 
most notable feature was the island's mountain, called 
R.hun (from Skr.) in the Akhbar al-Sin wa 'l-Hind (writ- 
ten 236/851), § 5 (a text much used by subsequent 
writers, e.g. al-Mas'udl, Murudj., i, 167-8 = §§ 175-6, 
etc.), regarded as the spot to which Adam descended 
on his expulsion from Paradise, leaving a footprint 
given as 70 cubits in size. At a later time, Ibn Battuta 
was to note that Muslims, Hindus and Buddhists alike 
regarded the mountain as holy, with the Buddhists 
considering the footprint as that of the Buddha (Rihla, 
iv, 165-85, German tr. H. von Mzik, Die Reise des 
Araber Ibn Batu.ta dutch Indien und China (14. 
Jahrhundert), Hamburg 1911, 353-67). 

The Akhbar further states (§§ 5, 51) that Sarandlb 
had two kings and that it was the custom of the island 
that a dead king was immolated with his wives also 
throwing themselves on the funeral pyre, if they wish- 
ed. Sauvaget observed that the detail of the two kings 
was known from the time of Cosmas Indicopleustes 
(6th century A.D.), reflecting the frequent division of 
power between a Singhalese monarch and a Tamil 
one in the north-east of Ceylon. Also mentioned are 
the import of wine from 'Irak and Fars by Muslim 
merchants for the king to enjoy personally and to sell 
to his subjects, and, apparently, the existence in the 
forests of Ceylon of the .aboriginal Vedda people. 
Bibliography: See especially, J. Sauvaget, 
Mhbar as-Sin wa l-Hind. Relation de la Chine et de I'ln- 
de, Paris 1948, §§ 5, 51 and comm., and G. Fer- 
rand, Relations des voyages et textes geographiques ... 
relatifs a I'Extreme-Orient du VIII' au XVIII' siecles, 
Paris 1913, see index s.v. Sirandib, with trs. of the 
relevant texts, to which may be added Hudud al- 
c dlam, tr. Minorsky, 61, § 5.2, comm. 194, and 
Marvazi, tr. idem, 46, 50. For Idnsi's detailed ac- 
count of Sarandlb and his map, see S. Maqbul 
Ahmad, India and the neighbouring territories in ... al- 
Sharif al-Idrisi, Leiden 1960, text 27-30, comm. 122- 
6 and Map II. For the spread of Islam there and 
subsequent history of the island, see ceyi.on. 

(C.E. Bosworth) 
SARANGPUR, a small town in Central India, 
before Partition in the Native State of Dewas, now in 
the Shajapur District of the state of Madhya Pradesh 
in the Indian Union (lat. 23° 34' N, long. 76° 24' E). 
It is essentially a Muslim town, founded by the sultans 
of Malwa [q.v.\, but on an ancient site. It was 
reputedly the location of a battle in 840/1437 when 
Mahmud Khaldji I of Malwa was defeated by the 
forces of Mewaf [q.v.], and, of more certain historici- 
ty, it was captured in 932/1526 from Mahmud II of 

Malwa by Rana Sanga [q.v.] of Citawr. Then in 
968/1561 it was seized by Akbar from the local gover- 
nor Baz Bahadur and incorporated into the Malwa 
suba of the Mughal empire, becoming the chef-lieu of 
the sarkar of Sarangpur. After falling to the Marathas 
[q.v.] in 1734, it was in 1818 restored to the Dewas 
State. The town contains several ruinous Islamic 
buildings, including a mosque with an inscription 
from 901/1496 by Ghiyath al-DIn Shah Khaldji. 

Bibliography: Imperial gazetteer of India 2 , xxii, 95- 
7; K.A. Nizami and M. Habib (eds.), A comprehen- 
sive history of India. V The Delhi Sultanate (A. D. 1206- 
1526), Delhi, etc. 1970, 790-1, 936-7. 

(C.E. Bosworth) 
SARAPARDA (p., lit. "palace-curtain"), the term 
applied in the sources for the Great Saldjuks and the 
Rum Saldjuks to the great tent carried round by the 
sultans, regarded, with the c'atr or mizalla [q. v. ], as one 
of the emblems of sovereignty. It is described in such 
sources as Rawandl, Rashld al-Din and Ibn Bibi as 
being red, the royal colour, and as having internal 
curtained compartments forming rooms. 

Bibliography: I.H. Uzuncarsih, Osmanh devleti 
teskilatina medhal, Istanbul 1941, 31, 37, 121; 
Sukumar Ray, Bairam Khan, Karachi 1992, 232. 
al-SARAT (a. "the back"), the coUective name, 
not particularly widespread, of the chains of moun- 
tains which run from the Gulf of c Akaba down to the 
Gulf of Aden [see al-<arab, DjazFrat, ii]. The word 
sarat occurs quite often in the construct state, as in 
sardt al-azd, sarat al-hdn, etc. In both Saudi Arabia and 
in Yemen, al-Sarat separates the lowlands along the 
Red Sea [see al-ghawr; tihama] from the high 
plateau. The commonest view in the Arab sources is 
that al-Sarat is identical with al-Hidjaz [q.v.] "the 
barrier". As a whole, the chains of mountains are cut 
up into large and small ranges which intersect in all 
directions. Al-Sarat is in general treeless, with black 
rocky ravines, ridges, peaks and pinnacles. Break- 
neck paths and bridle paths, often hardly traceable on 
the rock, lead up to narrow gates which give access to 
mountain villages found on almost inaccessible 
heights. There are well-cultivated fields in terraces 
along the slopes and in the valleys, protected by a wall 
of large stones. The fields yield coffee, protected from 
the heat of the sun by shade-giving trees, grapes and 

The heat on the western slopes is tropical, reaching 
from 23° C. in January-February to daytime shade 
temperatures of 38° C, and frequently of 49° C, in 

the Arabian Peninsula moving southwards and pro- 
viding some precipitation [see makka. 3. The Modern 
City. Floods]. At night, the temperature drops con- 
siderably, and the mountain tops are frequently 
covered with snow. In March and April some rains 
fall, normally torrential, while the rainy period lasts 
from the middle of June to the end of September. A 
further particularity of the climate of the western 
slopes are the Tihama fogs, called urnma or sukhaymanl. 
The climate of the eastern slopes is extremely dry, but 
the valleys, because of the rainy seasons, have a 
perennial water supply and show great wealth in fruit, 
cereals, plants and trees. 

There are only a few gaps in the al-Sarat chains: 
from Yanbu c al-Bahr to Yanbu< al-Nakhl, and via 
Badr Hunayn to Medina; from Djudda to Mecca; 
from al-Shukayya via al-Darb to Abha [q.v.]; in 
Yemen, from al-Hudayda [q.v.] to San'a 3 and from 
al-Mukha to Ta c izz [qq.v.]. 

Bibliography: Hamdani, Sifa, vi, Index 



Geographicus, 58; L. Forrer, Sudarabien nach al- 
Hamdani's " Beschreibung der arabischen Halbinsel" , 
Leipzig 1942, esp. 62 ff.; Bakri, Mutant, ed. 
Wustenfeld, ii, Fihrist, 26, s.v..; Yakut, Mu'djam, 
vi, Index 115, iv, 1020; Mardsid al-ittild c , ed. Juyn- 
boll 1853, 20-2; E. Glaser, Von Hodeida nach SanV, 
in Petermanns Mitteilungen, xxxii(1886), 43; R. Bayly 
Winder, Saudi Arabia in the nineteenth century, London 
1965; British Admiralty, A handbook of Arabia, Lon- 
don 1946. (A. Grohmann-[E. van Donzel]) 
SARATAN (a.), masculine substantive (pi. sardtin) 
denoting crustaceans (kishriyyat) in genera] and, 
more specifically, those which are collected for 
human consumption. The root s-r-t evokes, on the 
one hand, the notion of eating greedily and, on the 
other, that of running rapidly. The form saratdn serves 
as a substantive, also as a verbal noun and an adjec- 
tive; it is only the substantive which is considered in 
this article. Being applied to edible crustaceans 
(mahdra), it has undergone considerable distortions ac- 
cording to specific regions; thus the forms en- 
countered include sarta^dn, salta'an, salta'dn, salta'-un, 
saV-atdn and zaV-atdn. Alongside this term, each crusta- 
cean bears other names which often recur to denote 
other species. 

I. The dec 

nak 'ankabUt, 

(1) The spider crab (Maia sq 
durum, c ukrdysha, midja. 

(2) The squill-fish (Scyllarus latus) and the mantis- 
shrimp (Squilla mantis): zir al-bahr, istdkuzd al-raml. 

(3) The crab (Carcinus): saratdn al-bahr and all the 
distortions of saratan mentioned above, '■akrisha, 
kamarun, abu kamrun from the Spanish camaron, dj_anib, 
bu djnib, abu djalambu, man c . 

(4) The edible crab (Cancer pagurus or Carcinus 
maenas): umm djniba, khumkhum, risi. 

(5) The prawn and the shrimp (Penaeus caramote, 
Palemon serratus): abaydn, irbiyan, urbiydn, rubyan, bur- 
ghuth al-bahr, naylun, kumbri, kunbri, bu kusa, djambari, 
djammari, bu kamrun, kazzdzi. 

(6) The crayfish, river lobster (Astacus fluviatilis): 

tioned above, gharan, djardd al-nahr. 

(7) The lobster (Homarus vulgaris): saratdn al-bahr, 
zaPalan bahri, irbiyan, urbayan, rubyan, himar, 
bdbdslpdpds, karnit, kunbdr, karkand, bu makkusa, djardd 

(8) The crawfish, spiny lobster (Palinurus vulgaris): 
angust, ankush, istdkuzd 'l-shu'db al-murdjaniyya and all 
the names of the crab, the prawn, the crayfish and the 

(9) The hermit crab, soldier crab (Pagurus bernhar- 
dus): kata, saratan ndsik. 

2. Literature. 

Few ancient authors have discussed crustaceans. 
Aristotle, in his Historia Animalium (see Bibl.), offers 
some observations regarding the ethology of these 
aquatic creatures, observations which later Arab 
authors were content to reproduce to the letter; these 
include al-Djahiz, in the 3rd/9th century and, later, 
al-Damiri in the 8th/14th century, in his brief article 
saratdn (see Bibl.) concerning the crayfish and ancient 
legends relating to it. Thus finding in a village a dead 
crayfish, lying on its back, is a mark of protection 
against plagues and natural disasters. If it is attached 
to a fruit-tree, the latter will bear an abundant crop. 
In the Sea of China, a marine crustacean which comes 
ashore and dies of desiccation provides Chinese physi- 
cians with an ingredient for medicines designed to 
combat leprosy. In the same period as that in which 
al-Damiri was writing, Ibn Mangli included', in his 
treatise on hunting (see Bibl.), a brief chapter on the 

crayfish borrowed from his contemporary. It is in- 
teresting to note that for all these ancient authors the 
crustaceans are not decapods, but octopods, the pair 
of pincers (minkash, minkdkh, malkat) not being regard- 
ed as feet. 

3. Permissibility of consumption. 

By virtue of the Kur'anic verse (V, 95/96): "You 
are permitted the game of the sea (sayd al-bahr) and the 
food which is found there", crustaceans taken alive 
may, once cooked, be lawfully eaten. 

According to al-Damiri, the flesh of edible crusta- 
ceans is beneficial in the treatment of dorsal pains and 
of phthisis. Bearing on one's person the head of a 
crayfish prevents sleep when the moon is shining 
brightly, but induces sleep when there is no moon. If 
a crayfish is roasted and pulverised, the powder, ap- 
plied to haemorrhoids, causes them to subside. If the 
pincer of a crayfish is applied to a fruit-tree bearing a 
full crop, all the fruits will fall without the slightest 
cause. If a crayfish is applied to a deep wound enclos- 
ing an arrowhead, the latter is easily extracted. Final- 
ly, the crayfish serves as a talisman against any bite of 
a snake or a scorpion. 

5. Oneiromancy. 

Seeing a crustacean in a dream is the sign of a per- 
son of great guile, strongly armed, very preoccupied, 
going far afield in search of possessions and of un- 
sociable nature. One who dreams of eating the flesh of 
a crayfish could receive good things from a faraway 

6. Astronomy. 

Al-Saratan, "the Crayfish" or "Cancer" is the 
fourth zodiacal constellation containing the two stars 
known as "the two Pincers" (al-zubdnaydri), these 

(1) a (alpha) Cancri, mag. 4,4 (zubdna 'l-saratan al- 
djanubiyya) "the southern Pincer of the Crayfish"; and 

(2) t (iota) Cancri, mag. 4,2 (zubdna 'l-saratan al- 
shamdliyya) "the northern Pincer of the Crayfish". 

7. Medicine. 

Al-saratdn is the medical name currently given to the 
disease of cancer, with the synonym dkila, "the 
devourer", and the adjective saratdm, "cancerous". 
Bibliography (in alphabetical order of authors): 
Aristotle, Historia Animalium, tr. J. Tricot, Paris 
1957, i, 222-32; A. Benhamouda, Les noms arabes des 
etoiles (attempt at identification), in AIEO Alger 
(1951), ix, 147-9; Chenu, Encyclopedic d'histoire 
naturelle (vol. Crustaces), Paris 1877, 1-70; Damfrl, 
K. Haydt al-hayawdn al-kubrd, ed. Cairo 1937, s.v. 
saratan, ii, 19-20; Djahiz, K. al-Hayawan, Cairo 
1947, s.v. saratan and passim; Ibn Mangli, Uns al- 
mald..., Delachasse, tr. F. Vire, Paris 1984, 225-6; 
A. Malouf (al-Ma c luf), Mu c djam al-hayawdn. An 
Arabic zoological dictionary, Cairo 1932, s.v. saratdn, 
47, 74-5, 152; P. Kunitzsch, Arabische Sternnamen in 
Europa, Wiesbaden 1959, 205 no. 179; H. Eisen- 
stein, Einfuhrung in die arabische Zoographie, Berlin 
1991, Tier-Index s.v. "Krebs"; G. Oman, L'lch- 
tyonomie dans les pays arabes (Red Sea, Indian Ocean, 
Persian Gulf) (in Italian, English, French), publ. 
Instituto Universitario Orientale, Naples 1992, 
129-31; A. Salonen, Die Fischerei im alien Mesopota- 
mien nach sumerisch-akkadischen Quellen, in Annates 
Academiae Fennicae, Helsinki 1970, 156-7, 210. 

(F. Vire) 
SARAWAK, a state on the west coast of the 
island of Borneo and a constituent part of the 
Federation of Malaysia since 1963. Originally the 
name referred to a dependency of the sultanate of 
Brunei consisting of the Sarawak, Samarahan and 


Lundu river basins. Through a series of treaties (the 
first in 1841) with the Sultan of Brunei [q.v. in 
Suppl.], these territories passed to the "White Ra- 
jahs", the Brooke dynasty who administered Sarawak 
between 1841 and 1946. In 1946 the Brookes ceded 
their territory to the British Crown, and Sarawak, 
together with Sabah [q.v.], became British colonies. 
The wide variety of indigenous ethnic groups includes 
Ibans, Bidayuh (Land Dayaks), Melanau, Kayans, 
Kenyahs, Klemantans, Muruts and Malays. The 
general term "Dayak" was widely used to describe 
any non-Muslim peoples. 

The Muslim population has generally been held to 
be about 20% of the total and is made up primarily 
of Malay-speaking riverine and coastal dwellers. 

Brooke policy was to give effect to native laws and 
customs, and Islam was but one amongst a number of 
recognised law systems. There was no attempt to ad- 
minister the strict principles of the Shari^a until very 
recently (see below). Instead, it became the practice to 
note down the main principles of Islam as these were 
seen to affect public administration, and to enforce 
them through administrative procedures. Marriage, 
divorce, inheritance and conversion were all regulated 
in this way. 

From the early years of this century, Muslim mat- 
ters were governed by the Undang- Undang Mahkamah 
Melayu Sarawak (Laws of the Sarawak Malay Court). 
This document was a compendium of Malay custom 
and amended from time to time. It never constituted 
a "Code of Muslim Law" but instead was directed 
toward the regulation of a society whose members 
shared recognisably Muslim values, particularly as to 
relations between the sexes. No hard and fast line was 
drawn between custom (adat) and the Ska 

it the o 

:t betw. 


tional and modernist Muslims which occurred 
Peninsular Malay states was absent in Sarawak, 
possibly because the small number of local religious 
scholars studied in Mecca, not in Cairo, the centre of 
late 19th century reformist ideas, and because Brooke 
rule isolated Sarawak Muslims from contact with the 
international Muslim community. Also, the pondok 
schools [see pesantren] which on the Peninsula pro- 
vided elementary training in the Muslim sciences, 
were not developed in Sarawak. In 1939 a group of 
Malays in Kuching established the Persatuan Melayu 
Sarawak (the Sarawak Malay 

:t Islan 


banned under the Japanese Occupation of Sarawak, 
although there was otherwise little interference by the 
Japanese in the everyday affairs of Malays (Muslims). 
This rather static position has been fundamentally 
changed from the 1980s. A Department of Religious 
Affairs has been established, Muslim officials have a 
defined status in the administrative system of the 
state, a Sharf-a Court system is in place and hadjdi and 
educational finances are provided. Islam has also a 
political voice in state affairs as in the other states of 
Malaysia [q.v.]. Dakwah activity has been increasingly 

Bibliography: R. Pringle, Rajahs and rebels: the 
Ibans of Sarawak under Brooke rule, 1841-1941, Lon- 
don 1970; J. P. Ongkili, Pre-Western Brunei, Sarawak 
and Sabah, in Sarawak Museum Journal, N.S. xx/40-1 
(1972); M. Leigh, The rising moon: political change in 
Sarawak, Sydney 1974; C. Lockhard, The Southeast 
Asian town in perspective: a social history of Kuching, 
Malaysia, 1820-1970, 2 vols., Ann Arbor 1974; 
M.B. Hooker, Native law in Sabah and Sarawak, 
Singapore 1980; R.H.W. Reece, The name of Brooke. 
The end of White Rajah rule in Sarawak, Kuala Lum- 

pur 1982; M.B. Hooker, Islamic law in South-East 

Asia, Singapore 1984. 

(Virginia Matheson Hooker) 

SARAY, or Sarai, the name of two successive 
capitals of the Golden Horde located on the lower 
Volga. Of Persian origin, saray "palace" or "court", 
entered Turkic in the 1 1th century, where it was often 
paired with karshi, from the Tokharian B kerciye "royal 
palace", to designate the principal camp of a nomadic 
ruler (Nadelyaev, et ai, Drevnetyurskiy slovar' , 429, 
488 and Clauson, Etymological dictionary, 664). In the 
Mongolian era, 13th- 14th centuries, there was a fur- 
ther proliferation of Sarais (see, for example 
Galstyan, Armyanskie istocniki, 28) and this has given 
rise to considerable uncertainty and debate over the 
nomenclature, locale and chronology of the Golden 
Horde capitals. 

The first of these capitals, now called Old Sarai for 
the sake of clarity, is located along the left bank of the 
Akhtuba, an eastern tributary of the Volga, near the 
modern village of Selitrennoe, about 125 km/77 miles 
north of Astrakhan. Founded by Batu, the son of 
Djoci, Cinggis Khan's eldest son, some time after the 
conquests of Russia and the Kipcak steppe (completed 
around 1242), Sarai is first mentioned by name in 
1254 when the Franciscan William of Rubruck visited 
the site on his way to Mongolia. He relates only that 
Sarai was newly-built by Batu and had a palace 
(Mongol mission, ed. Dawson, 207, 210). The character 
of the new capital is nicely evoked in Djuwaym's 
depiction of Sarai as both a "camp" (mukhayyam) and 
as a "city" (shahr) (i, 222; Djuwayni-Boyle, i, 267). 
According to the archeological evidence, Sarai was 
a large complex, about four km/12!4 miles in length 
and covering an area of 10 km 2 . The city had a well- 
developed network of streets (usually defined by 
drainage ditches), water reservoirs, markets and 
substantial artisans' quarters which housed jewellers, 
metalsmiths and glass- and ceramics-makers. The 
walled villas of the wealthy were generally located on 
the outskirts of the city (Egorov, Isloriceskaya 
geografiya, 114-117; Fyodorov-Davydov, Golden Horde 
cities, 19-22). 

The most informative literary sources on Old Sarai 
all date to the 1330s, just before the move of the 
capital upstream to New Sarai (Saray al-Djadid), and 
consequently there has always been some dispute over 
which city is described in these accounts. In most 
cases, however, a careful reading of the source 
resolves the apparent ambiguity. Ibn Battuta, for ex- 
ample, begins his description of "al-Sara" by stating 
that it took four days of travel to reach the capital from 
Hadjdj Tarkhan. the modern Astrakhan. Clearly, 
therefore, he visited Old Sarai, which is 125 km/77 
miles from Astrakhan, and not New Sarai, which is a 
further four days' journey to the north. Moreover, the 
dimensions of the Sarai he depicts also confirm this 
conclusion; his city, like Old Sarai, is large, a half- 
day's ride in length (tr. Gibb, ii, 515-16), while the 
New Sarai, according to recent topographical studies 
(see below), was much smaller, barely 2 km 2 in area. 
Other contemporaneous accounts, by Abu 'l-Fida 3 
and al- c Umari, also picture a large, populous city and 
are, in all probability, referring to Old Sarai (Abu '1- 
Fida> Takwim, tr. ii, i, 322-3; al- c Umari, ed. Lech, 
Ger. tr. 146, Arabic text 81). 

The city described by these authors was full of mer- 
chants and markets. European sources, such as the 
commercial manual of Pegolotti, correctly place Sarai 
on the main overland trade route leading from the 
Crimea to China (Yule, Cathay, iii, 147; cf. Bratianu, 
Recherches, 239-41). The inhabitants, as was typical of 

Mongolian capitals, were diverse: Muslims and 
Christians and numerous ethnic groups including 
Alans, Circassians, Russians, Kipcaks and Greeks, 
each of which, according to Ibn Battiita, had its own 
quarter and bazaar. In its early decades Sarai had a 
substantial Christian, mainly Orthodox, population. 
In 1261 a bishop was appointed to Sarai to minister 
to its Christian residents and visitors. This bishop also 
played a central role in diplomacy between the 
Mongols, the Russian principalities and Byzantines 
{Nikon chronicle, tr. Zenkovsky, iii, 37, 45, 63, 132; 
Meyendorf, Byzantium, 45, 46, 78, 132, 150, 185). 
The Orthodox presence at Sarai is reflected in the 
numerous finds of metal icons and other religious ob- 
jects (Poluboyarinova, Russkie lyudi, 49-54). Over the 
course of time, and particularly during the early 
decades of the 14th century, the city took on a more 
Muslim character. Ozbek, the khan of the Golden 
Horde (r. 1311-41), and a convert to Islam, built a 
madrasa in Sarai and successfully attracted Muslim 
scholars to his capital (al- c Umari, ed. Lech, Ger. tr. 
136, Arabic text 68). By the late 1330s when the Fran- 
ciscan Pascal of Vittoria reached "Sarray", he viewed 
it as "a city of Saracens", and notes further that 
several years prior to his arrival Christian mis- 
sionaries were martyred there (Yule, Cathay, iii, 82, 

The transference of the capital from Old to New 
Sarai took place, by the best available evidence, at the 
end of Ozbek's reign. The issue is complicated, how- 
ever, by the existence of contradictory information on 
the origin and name of the new capital. Muslim 
writers of the 1 4th and 1 5th centuries often refer to a 
second Sarai founded by Berke (r. 1257-66), the first 
ruler of the Golden Horde to embrace Islam. Accord- 
ing to a tradition related by Ibn c Arabshah, the 
historian of the Timurid period, Berke first con- 
structed and then peopled his Sarai with Muslim 
divines as a conscious means of spreading Islam in the 
pagan steppe (tr. Sanders, 77-9; cf. al- c Umari, ed. 
Lech, Ger. tr. 146, Arabic text 81). Moreover, 
Natanzf, writing in the early 15th century, refers to 
both a Saray-i Barka and a Saray-i Batu (ed. Aubin, 
81, 97, 366). Such terminology, however, is never en- 
countered in earlier sources. Rashld al-Dln, for exam- 
ple, speaks of Batu's death at Saray, and more 
significantly, of Berke's burial at Saray-i Batu (ed. 
Karlml, i, 122, ii, 744). Berke's capital, an island of 
the faith in a sea of paganism, is, in all likelihood, a 
pious fiction, one that has bedevilled many 19th- and 
20th-century scholars who have sought to equate 
Saray-i Barka with Saray al-Djadid. In reality, Saray 
al-Djadld was built in the 1330s and became the new 
capital in the early 1340s when Djanibek (r. 1342-57) 
came to power. Several lines of evidence point to this 
conclusion. First, the earliest literary reference to 
New Sarai is a report of an anonymous 14th-century 
Mamluk author who relates that Ozbek died there in 
1341 (Tizengauzen, Sbornik materialov, i, Russian tr. 
263, Arabic text 254). Second, and even more helpful, 
is Ibn c Arabshah's statement that "between the 
building of Saray and its devastation there passed 
sixty-three years" (tr. Sanders, 79). Since its destruc- 
tion by Tfmur occurred in 1395, the city was founded 
in ca. 1332. The numismatic evidence also supports 
this chronology; minting, which began in Old Sarai 
around 681/1282-3, is drastically reduced after 
740/1339-40, and in the following year, 741/1340-41, 
silver dirhams appear for the first time in Saray al- 
Djadid. Thereafter, Saray al-Djadld becomes one of 
the most active mints in the Golden Horde, while Old 
Sarai's output is limited and intermittent (Fedorov- 

Davidov, nacale monetnoy cekanki, 83; Mellinger, 
Coins of the Golden Horde, 170-3). Lastly, extensive ar- 
cheological investigation of the site shows that it was 
founded in the 14th century; no evidence has emerged 
to support an earlier date. 

New Sarai, modern Tsarev, was located on the 
Akhtuba. just below the great bend of the Volga, 
about 125 km/77 miles north of Old Sarai. The area 
encompassed by New Sarai is well defined by a defen- 
sive ditch that encircled the city. Oval in shape, New 
Sarai was 1 .6 km in length and 1 km wide. Inside are 
the remains of a modest earthen embankment with 
strong points guarding the main gates. In the view of 
the excavators, the ditch and the wall were not part of 
the original construction but a later addition, proba- 
bly dating to the 1360s. Within the walls, the New 
Sarai, like the Old, had a network of streets defined 
by drainage ditches, along which were located in- 
dividual homes. In the southeastern quarter of the 
city, there were a number of fenced-in villas, while the 
poorer classes and artisans were concentrated in the 
centre of the city where they often lived in earthen 
dugouts {zemlyanka in Russian). Northwest of the 
main city was a suburb with its own walls, streets and 
water supply (Fedorov-Davidov et alii, Arkheologiceskie 
issledovaniya, 68-71). The living quarters unearthed at 
New Sarai reveal, not unexpectedly, an admixture of 
styles. The square and rectangular houses of wood 
and brick find their prototypes in Kh w arazm. the 
well-attested use of yurts in the city goes back, of 
course, to Mongolian-Turkic traditions, while the 
earthern dugouts show continuity with the local 
Saltovo-Majaki culture of the 5th-10th centuries 
(Egorov, Zilishca, 172-93). 

The internal history of the new capital is little 
known. One can infer that it was touched but not 
devastated by the plague which repeatedly swept 
through Russia and the Volga region between 1345- 
54 {Nikon chronicle, ed. Zenkovsky, iii, 157). This 
threat, as Mellinger has suggested, might explain the 
appearance of new coins in 749/1348-49 bearing the 
inscription Saray al-Mahrusa, "Saray the preserved 
[of God]" {Coins of the Golden Horde, 178-80). New 
Sarai, from the number of metal, stone, and ceramic 
crosses found there, also had a Christian population, 
and it seems likely that the "bishops of Sarai" ap- 
pointed in the latter half of the 14th century, because 
of the continued importance of their diplomatic func- 
tions, were now stationed at New Sarai, near the court 
of the Golden Horde (Poluboyarinova, Russkie lyudi, 
54-72; Nikon chronicle, ed. Zenkovsky, iii, 242, iv, 14, 
135, 139). 

New Sarai was buffeted by the growing turmoil 
within the Golden Horde. Under Mamai, a non- 
Cinggisid general who attempted to control the Horde 
through puppet rulers, the capital became a major 
centre of resistance to his rule. In 1361, according to 
a Russian source, "the Lords of Sarai" rebelled, 
"fortified Sarai", elevated a khan of their own, and 
in the following year fought a battle with the forces of 
Mamai {Nikon chronicle, ed. Zenkovsky, iii, 189). Ap- 
parently it is these events that produced the ditch and 
embankment uncovered by Soviet archeologists at 
New Sarai. Mamai later met defeat at the hands of the 
Moscovite Principality in 1380 and Toktamish, a 
Cinggisid from the eastern wing of the Djo£id line, 
seized control of the Golden Horde with the support 
of Tfmur. According to Russian and Turkish sources, 
he occupied "Sarai" where he was formally enthron- 
ed {Pamyatniki literaturi drevney Rusi, 191, 192; Abu '1- 
Ghazi Bahadur Khan, Histoire, tr. Desmaisons, 171). 
That New Sarai is probably meant here is supported 

by the fact that Toktamish issued considerable 
coinage at Saray al-Djadld throughout his reign. But 

measure of ambiguity remains (Mukhamadiev, 
Monetnaya sistema, 147, 152). In this connection, how- 
ever, it is important to consider that for the Mongols 
the concept of a single and fixed capital had far less 
significance and meaning than it did for sedentary 
peoples and that they continued their nomadic life- 
style long after the empire was founded. The elder 
Polos, for example, met Berke in the vicinity of 
Oucaca (Ukek), a summer pasturage halfway between 
Sarai and Bulghar, and Russian sources indicate that 
Ozbek had a camp as far west as the Sea of Azov 
(Marco Polo, ed. Moule and Pelliot, i, 74, 76; Nikon 
chronicle, tr. Zenkovsky, iii, 104). This pattern per- 
sisted into the later 14th and 15th centuries, for the 
extant yarliks (decrees) of the Golden Horde clearly re- 
veal that its rulers were still on the move, making an- 
nual rounds (Usmanov, Zalovannle aiti, 264-5). For 
the Mongols, their successive capitals, Old and New 
Sarai, were much enlarged winter camps which hous- 
ed them for only part of the year (see al- c Umari, ed. 
Lech, Ger. tr. 147, Arabic text 83; cf. also Gyorffy, 
Systeme des residences, 48-53, 135). This is why the Rus- 
sian chroniclers, who report endless official trips to the 
Khan, always say that their prince went to the orda, 
never to Sarai, for they well understood that the actual 
"capital" of the Horde was wherever the moving 
camp (orda) of the Khan was located at a given point 

The rapid decline of the Golden Horde, quite visi- 
ble by the last decades of the 14th century, had im- 
mediate and disastrous repercussions for all the 
Golden Horde cities of the Volga basin. When 
Toktamish and his patron TTmur fell out, the latter 
launched a major punitive campaign which 
culminated in the winter of 1395 with the destruction 
of Hadjdj Tarkhan and New Sarai, or Sarai the Great 
(Saray Velikiy) as the Russian "Story of Temir 
Aksak" has it (Pamyatniki literaturi drevney Rusi, 232, 
233). According to Timurid historians, Sarai and its 
surrounding districts were sacked, levelled and set 
ablaze (Sharaf al-DIn < All Yazdl, ed. 'AbbasI, i, 552; 
Nizam al-DIn ShamI, ed. Tauer, i, 164). Yet despite 
the claims of total devastation, these cities were at 
least partially rebuilt. In the reign of the Khan 
Muhammad (1421-45) Hadjdj Tarkhan, Saray al- 
Djadrd and Sarai were again issuing a limited number 
of coins (Mukhamadiev, Monetnaya sistema, 157). 

The final destruction of the Golden Horde capitals 
can be dated to the last half of the 15th century. 
Afanasiy Nikitin, a Russian merchant who traversed 
the entire length of the Volga in 1466, mentions only 
one Sarai in his detailed itinerary (Nikitin, Khozenie, 
1 1 , 34, 53, 71). By this time, evidently, one of the two 
capitals had already faded into obscurity. And the sur- 
vivor, whatever its identity, was soon under attack. In 
1471 troops from the Russian town of Vyatka raided 
down the Volga, temporarily occupied Sarai, seized 
much booty, and successfully returned home 
(Iosafovskaya letopis' , ed. Zimin, 73). Another Russian 
raid came in 1480, and in the following year the 
Siberian and Noghai Tatars joined forces and 
systematically pillaged all the camps of the Great 
(Golden) Horde "between the Don and Volga" 
(Kazanskaya istoriya, ed. Moiseeva, 56; Ustyuzskii 
letopismi svod, ed. Serbina, 93-4). The Golden Horde 
had been dealt a crippling blow from which it never 
recovered and Sarai, its long-time capital,- departs 
from the historical stage. 

Well-known and important cities in the Islamic 

world and Eastern Europe, the fame of the Sarais 
spread much further afield. Chaucer, in The Squire's 
Tale refers to "Sarray in the Land of Tartarye" and 
Sarai/Sarra appears on European maps in the 1320s 
with anachronistic references as late as the 17th cen- 
tury. Fra Mauro's world map of 1459 even registers 
the two Sarais, the "great" and the "small" (Tardy, 
Contribution, 182, 184, 185-6, 189, 190, 206, 212, 

Bibliography: 1. Sources, (a) Arabic. Ibn 
c Arabshah, Tamerlane or Timur the Great Amir, tr. 
J.H. Sanders, repr. Lahore 1976; Abu '1-Fida', 
Takwim al-bulddn, tr. Reinaud, Geographic 
d'Aboulfeda, ii/1, Paris 1848; Ibn Battuta, Travels, ii, 
tr. H.A.R. Gibb, Cambridge 1962; V. 
Tizengauzen, Sbornik materialov olnosyashcikhsya k 
istorii Zolotoy Ordi, i, Izvleceniya iz socineniy arabskikh, 
St. Petersburg 1884; c UmarI, Das mongolische 
Wellreich, ed. and tr. K. Lech, Wiesbaden 1968. (b) 
Turkic. Abu '1-GhazI Bahadur Khan, Histoire des 
Mongols et des Tatares, tr. P.I. Desmaisons, repr. St. 
Leonards and Amsterdam 1970. (c) Persian. Dju- 
waynl, Ta\ikh-i djihan-gusha, ed. Kazwlni, i; idem, 
i, tr. J. A. Boyle; Mu c In al-DIn Natanzl, Muntakhab 
al-tawdrikh-i muHni, ed. J. Aubin, Tehran 
1336/1957; Rashld al-DIn, Dj_am? al-tawdrikh, i, ii, 
ed. B. Karlml, Tehran 1338/1959; idem, The suc- 
cessors of Genghis Khan, tr. J. A. Boyle, New York 
1971; Nizam al-DIn ShamI, Histoire des conquetes de 
Tamerlan intitulee Zafar-nama. i. Texte persane, ed. F. 
Tauer, Prague 1937; Sharaf al-DIn c Alf Yazdl, 
Zafar-nama, i, ed. M. c AbbasI, Tehran 1336/1957. 
(d) Russian. Iosafovskaya letopis', ed. A. A. Zimin, 
Moscow 1957; Kazanskaya istoriya, ed. G.N. 
Moiseeva, Moscow-Leningrad 1954; Afanasiy 
Nikitin, Khozenie za tri morya, 2 Moscow-Leningrad 
1958; The Nikon chronicle, iii-iv, tr. S. Zenkovsky, 
Princeton 1986-8; Pamyatniki literaturi drevney Rusi: 
XlV-seredina XV veka, Moscow 1981; Ustyuzskii 
letopisnii svod, ed. K.N. Serbina, Moscow- 
Leningrad 1950. (e) Armenian. A.G. Galstyan, 
Armyanskie istocniki o Mongolakh, Moscow 1962. (f) 
European. Marco Polo, The description of the world, i, 
ed. A.C. Moule and P. Pelliot, London 1938; 
Pascal of Vittoria, Letter, in Yule, Cathay and the way 
thither, iii, repr. Taipei 1966; Francesco Balducci 
Pegolotti, Lapractica della mercatura, in Yule, op. cit. , 
iii; William of Rubruck, Journey, in C. Dawson 
(ed.), The Mongol mission, New York 1955. 




Egorov, Zilishca Novogo Saray, in A. P. Smirnov 
(ed.), Povolz'e v srednie veka, Moscow 1970, 172-93; 
G.A. Fedorov-Davidov et alii, Arkheologiceskie 
issledovaniya Tsarevskogo gorodishce (Noviy Saray) v 
1959-1966 vv., in Smirnov (ed.), Povolz'e v srednie 
veka, 68-171; idem, O nacale monetnoy cekanki v 
Khprezme i Saray v kontse XIII v., in Epigrafika 
vosloka, xiv, Moscow-Leningrad 1961, 79-89; G.M. 
Mellinger, The silver coins of the Golden Horde: 1310- 
1358, in Archivum Eurasiae MediiAevi, vii (1987-91), 
153-211; A.G. Mukhamadiev, Bulgaro-Tatarskaya 
monetnaya sistema XII-XV vv., Moscow 1983; S.A. 
Yanina, Monety Zolotoy Ordi iz raskopok i sborov 
povolzskoy arkheologiceskoy ekspeditsii na Tsarevskom 
gorodishce v 1959-1962 gg., in Smirnov (ed.), Povolz'e 
v srednie veka, 194-218. 

3. Studies. G.I. Bratianu, Recherches sur le com- 
merce ge'nois dans la Mer Noire au XIII' siecle, Paris 
1929; Sir G. Clauson, An etymological dictionary ofpre- 
thirteenth century Turkish, Oxford 1972; V.L. Egorov, 
Istoriceskaya geografiya Zolotoy Ordi v XII-XIV vv., 
Moscow 1985, 112-17; G.A. Fyodorov-Davydov, 


The culture of the Golden Horde cities (British Ar- 
cheological Reports, 198), Oxford 1984, 16-25; B. 
Grekov and A. Iakoubovski, La Horde d'Or et la 
Russie, Paris 1961, 135-47; Gy. Gy6rffy, Systeme de 
residences d'hiver et d'ete chez les nomades et les chefs 
hongrois de X' Steele, in Archivum Eurasiae Medii Aevi, 
i (1975), 45-153; J. Meyendorf, Byzantium and the 
rise of Russia, Cambridge 1981; V.M. Nadel^aev et 
alii, Drevnetyurkskiy slovar', Leningrad 1969; M.D. 
Poluboyarinova, Russkie ljudi o Zolotoy Orde, 
Moscow 1978, 49-72; B. Spuler, Die Goldene Horde, 
'Wiesbaden 1965, 264-70; J. Tardy, A contribution to 
the cartography of the central and lower Volga, in A. 
Rona-Tas (ed.), Chuvash studies, Wiesbaden 1982, 
179-236; M.A. Usmanoy, Zalovannie akti dzucieva 
ulus, Kazan 1979; B.B. Ziromirskii, Sarai, in The 
Modern Encyclopedia of Russian and Soviet History, 
xxxiii (Gulf Breeze, Fla. 1983), 101-2. 

(T.T. Allsen) 
SARAY (p.) (from an Old Persian form 'srdda, root 
0rd "to protect") means in Persian dwelling, 
habitation or house. The word is frequently com- 
pounded with another substantive to indicate a par- 
ticular kind of building. The best known example is 
kdrwan sardy "caravanserai", a roadside stopping- 
place for caravans [see khan]. Similarly, the Djannat- 
sardy added to the northern part of the shrine at Ar- 
dabll by the Safawid Shah Tahmasp I ca. 947/1540 is 
a domed octagonal building used for Sufi gatherings 
and prayer (A. H. Morton, The Ardabil shrine in the reign 
ofShih Tahmasp I, in Iran, xii [1974], 31-64, and xiii 
[1975], 39-58, no. C). The word sardy was extended to 
refer to the seat of government and the residence of a 
prince. It could refer to a town. Ak Saray [q.v.], for 
example, is a town in inner Anatolia with a castle built 
by the Saldjuk sultan of Rum, Kilidj Arslan II (r. 551- 
88/1156-92). Seray Berke [q.v.\, capital of the Blue 
Horde in southern Russia, takes its name from its 
founder Berke (r. 655-65/1257-67). The most com- 
mon meaning of saray, however, is palace, and this 
article is a general review of Islamic palaces. It in- 
cludes examples known by other names, but concen- 
trates on those in the eastern Islamic lands, where the 
word saray is common and on those that are not pub- 
lished elsewhere in the Encyclopaedia of Islam. 

In early Islamic times [see architecture], palaces 
were known in Arabic by various terms, including the 
rare balat [q.v. in Suppl.], derived from the Latin 

Umayyad desert palaces and frontier forts. Kasr al- 
Hayr al-Gharbl [q.v.], for example, was an Umayyad 
castle, and Kasr al-Hayr al-S_harki [q.v.], an 
Umayyad agricultural settlement. These fortified 
residences were square, two-storeyed structures, 
usually 35 or 70 m per side, with apartments of several 
rooms opening on to a central court. They are re- 
markable for their elaborate decoration, including 
floor mosaics, paintings, and carved stuccoes, and the 
fancier establishments formed part of larger com- 
plexes with mosques, baths, dams and other 
buildings. The biggest, the palace at Ukhaydir [q.v.] 
in southern c Irak, is generally considered to be some- 
what later (late 8th century) and is distinguished not 
only by its larger size (112 x 82 m) but also by its in- 
ventive vaulting and extensive decoration in carved 
stucco and brick laid in geometric patterns. Many of 
these residences can be considered to continue the 
Late Antique tradition of the villa rustica, centres of 
agricultural exploitation and private pleasure away 
from the prying eyes of the urban establishment. The 
square forms were repeated extensively in Central 
Asian palaces of the feudal aristocracy [see dihkan]. 

Their residences, known as kushk, were mud-brick 
buildings with a central court or domed hall surround- 
ed by living quarters. 

In contrast to these rural establishments from early 
Islamic times, many of which are still standing, con- 
temporary urban palaces are known primarily 
through texts and identified by other names. One was 
kubbat al-khadrb?. This was the name of the palace 
erected at Damascus by the first Umayyad caliph 
Mu'awiya (r. 41-60/661-80), and it reoccurs frequent- 
ly in early Islamic times, for example at the palace 
erected by the c Abbasids in their new capital at 
Baghdad, founded in 156/962. Although often 
translated as "Green Dome", referring to an oxidised 
copper coverning over a wooden dome, the term is 
better translated "Dome of Heaven", referring to the 
long-standing Mediterranean tradition of a heavenly 
dome over the ruler's throne. 

Another type of urban palace was the Dar al-Imara 
("House of Government"), usually built adjacent to 
the congregational mosque to allow the governor 
ready access. The Dar al-Imara erected by Sa c d b. 
Abl Wakkas [q. v. ] at Kufa in 1 7/638 just after the con- 
quest of c Irak was excavated between 1935 and 1956 
by the Iraqi Directorate of Antiquities. It was a square 
building (12, 100 m square, or 1 10 m on a side) enclos- 
ing a large courtyard with Twans on three sides and a 
triple-aisled hall and domed room on the south. The 
Dar al-Imara erected by Abu Muslim at Marw in 
129/747 is known through textual descriptions. It was 
also a square building, but had a central domed 
chamber surrounded by four Twans that opened on to 

With the founding of Samarra 3 [q.v.] by the 
'Abbasid caliphs in the 3rd/9th century, Islamic 
palaces changed dramatically and became significant- 
ly broader and lower. The gargantuan 176-ha palace 
founded by the 'Abbasid caliph al-Mu c tasim at 
Samarra' in 221/836 was known as the Dar al-Khilafa 
and later as the Djawsak al-Khakanl. Excavated by 
the German Samarra Expedition of 1911-13 and 
surveyed since 1983 by the Iraqi Directorate of Anti- 
quities, the site included immense gardens and two 
main palatial units. The southern one, used for public 
audience, was a square building_(180 x 200 m) with 
a monumental gateway (Bab al- c Amma), throne halls, 
rooms and a large courtyard on the east. The northern 
one, used for private residence, had a smaller square 
reception hall and residential apartments. The palace 
built by al-Mutawkkil two decades later was on a 
similar scale and had a corresponding division be- 
tween spaces for public audience and private 
residence. These sprawling palace-cities with Twans 
connecting closed and open spaces allowed visitors to 
process in an ascending crescendo towards the ruler, 
and the magnificent wall decoration, known from con- 
temporary descriptions and fragmentary remains, was 
intended to underscore the ruler's power and majesty. 

c Abbasid palaces set the model throughout the 
Islamic world in the mediaeval period, and other 
rulers copied not only the architectural units — two 
juxtaposed rectangular halls or four iwdns around a 
court— but also the sprawling spaces divided into 
public and private realms. The palace-city founded in 
325/936 by the Umayyads of Spain outside Cordova 
at Madinat al-Zahra 3 [q.v.], which has been excavated 
and reconstructed by the Spanish since 191 1, was laid 
out on the slopes of the Sierra Morena in a series of 
cascading terraces that took advantage of the site. 
From gardens in the lower zone, one ascended to 
workshops and court buildings in the middle zone and 
then to the Alcazar in the highest zone, which contain- 

ed the dar al-mulk, the private quarters of the caliph 
and his close associates. The large scale (the site 
measured some 1500 by 750 m or 1 12 ha) and layout, 
like the rich decoration, was meant to project the 
Umayyad caliph's role, as had the c Abbasid palace at 

The two palaces built by the Fatimid caliphs in their 
new city of al-Kahira [q.o.] were smaller but also 
splendid establishments. They flanked the main 
north-south road through the city, which became 
known as Bayn al-Kasrayn. The Great Eastern Palace 
built for the caliph al-Mu c izz (r. 341-65/953-75 [q.v.]) 
reportedly covered 10 ha and had 12 pavilions; the 
Small Western Palace built for his successor al- c Aziz 
(r. 365-86/975-996) was set in front of a vast garden. 
The Fatimid palaces were destroyed and the sites built 
over after the Ayyubids transferred the seat of power 
to the citadel. The palace and state buildings were 
located in the southern enclosure constructed in the 
second half of the 7th/13th century. The most im- 
pressive unit was the dome chamber of the Dar al- 
c Adl, built by the Mamluk sultan al-Nasir Muham- 
mad in 733/1333-4. It survived until the 19th century, 
when it was cleared to make way for the Mosque of 
Muhammad c Ali. The citadel, with palace and other 
establishments for the rulers, was a common feature 
of cities in the Levant during the mediaeval period 
which can be found in Damascus, Aleppo and 

Palaces in the eastern Islamic lands also repeated 
many of the same ideas established under the 
c Abbasids. Three mud-brick palaces were built at 
Lashkarl Bazar [q.v.], a royal suburb that extended 
six km northwards from Bust along the east bank of 
the Helmand River. Begun as a garden development 
in the 4th/ 10th century and expanded in the 5th/ 1 1th 
and 6th/12th centuries, the site was excavated by the 
French in the 1940s and 1950s (D. Schlumberger et 
alii, Lashkari Bazar. Une residence royale ghaznevide el 
ghoride, Mems. DAFA, xviii [1978]; see also the im- 
portant considerations by Terry Allen, Notes on Bust, 
in Iran, xxvi-xxviii [1988-90]). The earliest was the 
small Centre Palace, erected along the river bank with 
a nearby pavilion and large garden to the east. It may 
be the kushk erected by the Saffarid Tahir at the begin- 
ning of the 4th/10th century and is unrelated to the 
larger and better known South Palace, assigned to the 
reign of the Ghaznawid Mahmud (388-421/998- 
1030). In both plan— a large rectangle enclosing a 
central courtyard with twins on the four sides — and 
decoration — panels of stucco carved in relief — the 
South Palace at Lashkarl Bazar copies the Mesopota- 
mian prototypes used at the c Abbasid palaces at 
Samarra 3 . Its most distinctive feature is a second 
iwdn-ha\\ lying beyond the north twdn and overlooking 
the river. The grandest reception room in the palace, 
it had a dado painted with a frieze of attendants and 
walls revetted in baked brick and carved stucco in 
geometric patterns and inscription bands. The third 
palace at the site, the poorly-known North Palace, was 
a large rectangular enclosure containing three self- 
contained buildings, each with a central courtyard. 

The plan of twdns grouped around a courtyard, and 
rich decoration in carved stucco and painting, con- 
tinued to be hallmarks of palaces erected in the 
5th/llth and 6th/12th centuries at other sites in the 
eastern Islamic lands under later Ghaznawids and 
their local representatives. For example, the large 
palace erected by Mas c ud III (r. 492-508/1099-1115) 
to the east of Ghazna [q.o.] had a rectangular court 
(50 x 32 m) paved in marble and surrounded by 32 
niches and twdns on the four sides, with an additional 

throne room beyond the south twan. Excavations un- 
covered an extraordinary inscription in floriated 
Kufic that extends for 250 m around the court. The 
text is a Persian poem extolling the virtues of the 
sultan and the glories of his palace and was composed 
specifically for the new construction (A. Bombaci, The 
Kufic inscription in Persian oerses in the court of the Royal 
Palace of Mascud III at Ghazm, Rome 1966). Similarly, 
the palace erected by local rulers in the northern 
suburb of old Tirmidh [q. v. ] on the right bank of the 
Oxus River comprised several buildings grouped 
around a courtyard. Excavations by the Termez Ar- 
chaeological Expedition (TAKE; 1936-8) and others 
have yielded rich decoration in carved stucco, in- 
cluding panels with animals and other zoomorphic 
motifs. The palace built in the late 3rd/9th or early 
4th/10th century at Hulbuk, capital of the mediaeval 
province of Khuttalan [q.v.] on the Kyzyl (Akhsh) 
River in southern Tajikistan, was rebuilt three times, 
and each time the carved stucco panels and murals 
were renewed. The site, excavated since the 1960s, 
has yielded more than 5,000 panels and fragments of 
carved stucco, with an exceptionally diverse range of 
geometric, floral and epigraphic motifs (E. 
Gulyamova, Reznoy shtuk Khulhuka, Material 'naya 
kul'tura Tadzhikistana, iii [1978], 186-202). 

Nothing has survived of Saldjuk palaces in Persia, 
but they probably provided the models for the palace 
parks built by the Saldjuks of Rum in the 7th/13th 
century in or near Konya, Kayseri and Alanya. They 
had multi-storey pavilions commanding fine views, 
often over a nearby lake, and were resplendently 
revetted in glazed tiles. A riparian view also provided 
the setting for the contemporary Kara Saray ("Black 
Palace"), the palace built by the Atabeg Badr al-DIn 
Lulu 3 at Mawsil between 630/1233 and 657/1259 (F. 
Sarre and E. Herzfeld, Archdologische Reise im Euphrat- 
und Tigris-Gebiet, Berlin 1920, ii, 239-49). Although 
much of the palace was destroyed in the 1980s, it was 
a two-storey masonry construction with iwdns 
grouped around a central court and had fine decora- 
tion in carved stucco. It is one of the few examples of 
a saray in the Arab lands, but in its foundation inscrip- 
tion (RCEA, no. 4451) it was simply called bind'' 
("building" or "structure"), and the name Kara 
Saray is probably modern. 

With the advent of the Mongols, the tent, often set 
in a garden, became a major form of palatial architec- 
ture [see khayma]. These portable structures with 
rigid supports covered with fabric have not survived, 
but textual descriptions and depictions in later book 
paintings show that they were large and elaborately 
decorated ensembles. For example, the tent that 
Hiilegii used when he ascended the throne near Balkh 
was made of gold-on-gold material and attached by a 
thousand gold nails; it included an elevated pavilion 
and a magnificent audience hall decorated with gold 
and silver gem-studded vases (Rashid al-DIn, Djami" 
al-tdwdrtkh, ed. and tr. E. Quatremere, Histoire des 
Mongols de la Perse, Paris 1836, repr. Amsterdam 1968, 
159-65). The ruler's large tent supported by guy ropes 
(bdrgdh) was often combined with a trellis tent (khargdh) 
which served as a private chamber. These tents made 
fitting palaces for rulers: a trellis tent that belonged to 
the Ak Koyunlu ruler Uzun Hasan (r. 857-82/1453- 
78 [qv.]) had a wooden door painted in red and blue 
with scenes of fighting beasts and was covered with 
blue cloth and trimmed with red silk. 

The tent was not the only type of palace used in the 
later period, however, and other palaces built of more 


:rials c< 


. The large and elaborately decorated 

hunting lodge uncovered by German excavations in 
the early 1960s at Takht-i Sulayman, to the southeast 
of Lake Urmiya, was begun ca. 673/1275 by Abaka on 
the foundations of the Sasanid sanctuary of Shiz 
[q.v.], the site where the Sasanid emperors were 
crowned. The palace had an artificial pond in the cen- 
tre of a large courtyard (125 x 1 50 m) surrounded by 
porticoes and four iwdns and lavish decoration in carv- 
ed marble and lustre and ladjwdrdina; and a balcony 
supported on wooden columns afforded a fine view of 
the pond (R. Naumann, Die Ruinen von Tacht-e 
Suleiman und Zendan-e Suleiman, Berlin 1977). Textual 
descriptions and depictions in contemporary paintings 
show that other II Khanid palaces also used the iwdn 
plan and had elaborate decoration. By the Mongol 
period, the word saray was increasing applied to 
palaces, even those built earlier. The II Khanid 
historian Hamd Allah Mustawfi al-Kazwinl, for ex- 
ample, referred to the palace that the BuyTd c Adud al- 
Dawla (r. 367-72/978-83) had built in eastern 
Baghdad as saray-i sultan ("palace of the sultan"; 
TaMkh-i Guzida, ed. Nawa'i, Tehran 1339, 415). An 
enlarged and rebuilt version of the one his father 
Mu c izz al-Dawla had founded, the palace, which is 
totally destroyed, had a great court surrounded by 
domed porticoes, halls for audience, residences, and 

The TTmurids continued to use both the tent and 
the structural palace, and the word saray was also used 
with a qualifying adjective of colour to refer to a major 
palace or citadel. The most famous Timurid example 
is the Ak Saray ("White Palace"), the palace that 
Timur built between 781 and 798/1379-96 in his 
birthplace, Kish [see kash], later known as Shahr-i 
Sabz ("The Verdant City"). The palace was located 
in the northeast quadrant of the new walled enclosure, 
and its facade faced north towards Samarkand, the 
chief city of the realm. All that remains is the colossal 
entrance portal (pishtdk) with a vault 22 m wide, but 
from the description by the Spanish ambassador Ruy 
Gonzales de Clavijo, it is possible to reconstruct the 
plan of a courtyard building with four axial iwans. 
The name, which was used in contemporary sources, 
cannot refer to the actual colour of the building, which 
was resplendent with coloured tiles, but is thought to 
signify "aristocratic" (L. Golombek and D. Wilber, 
The Timurid architecture of Iran and Turan, Princeton 
1988, no. 39). The Gok Saray ("Blue Palace") that 
Timur built at Samarkand ca. 802/1400 was a four- 
storey palace that has been destroyed. Suburban 
palaces were often called bagh, meaning a park or 
estate with buildings and gardens; other terms used in 
the Timurid period include takht, meaning a pavilion 
with a view, khdna and manzil ("residence"), and kushk 
or kasr, some of which were larger than kiosks or 
pavilions (T. Allen, A catalogue of the toponyms and 
monuments of Timurid Herat, Cambridge, Mass. 1981, 
nos. 405 and 206) 

The palace precinct laid out between 998/1590 and 
1020/1611 by the Safawid Shah 'Abbas I in his new 
capital at Isfahan continues many Timurid traditions, 
especially the unfortified exterior and the garden set- 
ting dotted with pavilions, but most of the structures 
were built of masonry, and not fabric tents. The 
precinct, sandwiched between the Maydan-i Shah 
("Royal Square") and the Cahar Bagh, the boulevard 
with trees and water channels that leads south across 
the Ziyanda River, was divided into two zones. A 
more public zone near the maydan contained 
workshops and administrative offices, while a more 
private residential area behind it contained ga'rdens 
with small open pavilions. These were often known as 

hasht bihisht ("Eight Paradises"), from their plan — an 
octagon with eight rooms or apartments arranged 
around a large central hall. The form copies now- 
vanished Timurid models, known from other 
buildings such as the Qinili Kbsk ("Tiled Kiosk") 
added to the Topkapi Palace in Istanbul in 878/1473 
and the Hasht Bihisht, the palace that the Ak 
Koyunlu ruler Ya c kub built at Tabriz in 892/1486. 
The name was popular in the Safawid period, as in 
the pavilion built in Isfahan by Sulayman I/Safl II (r. 
1077-1105/1666-94), and many smaller examples. 
Columnar porches were another popular design for 
Safawid palaces, found in Isfahan at the Cihil Sutun 
and repeated at Khiva, where the Tash Hawli 
("Stone Courtyard") palace (1830-8) incorporated 
:ral pillared porticoes arranged around courtyards 


5 163 n 

of these Timurid and Safawid forms were 
copied in the palaces built under the Mughals in the 
Indian subcontinent, but they were incorporated 
inside walled palace-fortresses and executed in local 
materials, particularly red sandstone contrasted with 
white marble. In general, a main gate led to a long 
bazaar which gave on to the courtyard and hall for 
public audience. This public area was separated from 
a more private one that had courtyards with pavilions, 
often called mahall (q. v. for an extensive discussion of 
both pavilions and their setting; and see mughals. 7. 
Architecture, for a discussion of chronological and 
stylistic developments). Buildings were laid out to take 
advantage of topography and setting. At Fathpur 
Sikri [q.v.], Abkar's fiat city (979-93/1571-85), for ex- 
ample, the tallest building, the five-story pavilion with 
domed kiosk now known as the Pandj Mahall 
("Palace of Five [Levels]"), demarcates public space 
from private; in the Red Fort (1048-58/1639-48) built 
by Shah Qjahan in his suburb of Shahdjahanabad [see 
dihli. ii], the private pavilions overlook the Yamuna 
or Jumna River. The looser arrangement of earlier 
sites, such as the Red Fort at Agra, became increas- 
ingly regularised, as at Fathpur Sikri, and especially 
at Shahdjahanabad. This greater attention to sym- 
metry and axiality underscores the increasing role of 
ceremonial in Mughal life. 

Unlike the Safawids and Mughals, who built 
numerous palaces in several places, the Ottomans 
maintained their court in a single palace, the Saray-i 
humayun, founded by the Ottoman sultan Mehemmed 
II [q.v.] after the conquest of Constantinople in 
857/1453 and used as the primary residence of the Ot- 
toman sultans until the 19th century. Although com- 
monly known today as the Topkapi Seray after a 
shore pavilion built near a gate of that name, until the 
19th century it was called in Ottoman sources saray-i 
djedid-i ( dmire ("New Imperial Palace") or yenf saray 
("New Palace") to contrast it to the eski saray ("Old 
Palace"), the first palace that Mehemmed had built in 
the centre of city, and to the Tekfur Saray, the Byzan- 
tine palace (Giilru Necipoglu, Architecture, ceremonial, 
and power: the Topkapi Palace in the fifteenth and sixteenth 
centuries, New York and Cambridge, Mass. 1991, 4). 
Set on a magnificent promontory on the tip of the 
peninsula overlooking the Bosphorus and the harbour 
of the Golden Horn, the palace comprises an outer 
precinct or park, known as the First Court, and an in- 
ner precinct of three courts constituting the palace 
proper. A line of three great portals leading into the 
First, Second, and Third Courts provided a major 
ceremonial axis to the audience hall (T. W odast) set 
beyond the third portal. Behind were gardens with 
pavilions, and to the left were several residences with 
courtyards, including several for the sultan's harem. 


picked up in European languages, where several loan- 
words from sardy, such as the Italian seraglio and the 
French serail are sometimes found with the meaning 

The palace-city was also developed in the western 
Islamic lands, as at the Alhambra, built up in the 
7th/13th and 8th/14th centuries outside Granada [see 
gharnata. Monuments. B]. One of the best pre- 
served and most popular examples of an Islamic 
palace, it is actually a series of palaces, including the 
more public Comares Palace with its audience hall 
and the more private Palace of the Lions. 

The arrangement seen at the Alhambra reappears 
in palaces built later in the western Islamic lands. The 
ruined Badi c Palace, built at Marrakush [q.v.] by the 
Sa c dian ruler Ahmad al-Manstir (r. 986-1012/1578- 
1603) for official receptions, for example, is an in- 
flated version of the Court of the Lions. The biggest 
was the palace-city built by the Filall ruler Isma'il (r. 
1082-1139/1672-1727) at Miknas [q.v.], where seven 
km of thick mud-brick walls with bastions enclose 
three separate palaces, each comprising several 
smaller palaces of differing arrangements as well as 
pavilions, mosques, barracks, prisons, stables, 
granaries, olive presses and huge pools to water the 
extensive gardens (M. Barrucand, Urbanisme princier en 
Islam: Mekn'es el les villes royales islamiques posl-medievales , 
Paris 1985). The scale is far grander than at the 
Alhambra, and the site was so large that the palaces 
were never finished. Pise, made from the ever- 
available earth, was the main material of construc- 
tion, but other materials, such as columns and 
marbles, were salvaged from other sites or imported. 

In the 18th and 19th centuries, palaces in the 
Islamic lands were often built on European models or 
incorporated features of Western architecture, such as 
monumental stairways, tall windows, engaged 
pilasters, and landscape murals. Dolmabahce Palace, 
the Ottoman palace built on the shore of the 
Bosphorus in 1853-5 to replace Topkapi Palace, had. 
an opulent double staircase with rock crystal 
balustres, the first_ ceremonial staircase to survive 
since the Bab al- c Amma at Samarra 3 . The Kadjar 
dynasty of Persia continued the practice of seasonal 
migration that the II Khanids and TTmurids had 
maintained and had palaces throughout the country. 
Fath C A1I Shah (r. 1212-50/1797-1834) enlarged the 
Gulistan Palace in Tehran, a rambling series of 
buildings and gardens within a walled enclosure that 
served as winter residence and administrative centre, 
by adding the Takht-i Marmar, a columnar audience 
hall on the form of a traditional talar. To cater for his 
peripatetic lifestyle, he also established many summer 
palaces, often terraced structures set in elaborate 
gardens, including several in hillside villages subse- 
quently incorporated in the northern suburbs of 
Tehran. His palaces were often decorated with large 
murals depicting the Shah and his court. The largest 
was a mural painted by c Abd Allah Khan in 1812-13 
to decorate the Nigaristan Palace; it showed the ruler 
enthroned in state with 118 life-size figures. The spate 
of palace construction continued under Nasir al-DIn 
(r. 1264-1313/1848-96), but the elegant symmetry 
typical of Fath c Alr Shah's work was abandoned and 
European elements, such as pilasters and tall win- 
dows, were added to traditional features such as col- 
oured tilework, ornately carved stucco, and mirror 

Bibliography: In addition to the monographs on 
individual palaces cited in the text, see G.A. 
Pugacenkova, Puti razvitiya arkhitekturi yuzhnogo Turk- 

menistana port rabovladeniya ifeodalizma, Moscow 1958 
(esp. ch. 3 on the kushk of the 6th-10th centuries); 
K.A.C. Creswell, A short account of early Muslim ar- 
chitecture, Harmondsworth 1958, revised and sup- 
plemented by J.W. Allen, Aldershot 1989; O. 
Grabar, The formation of Islamic art, New Haven 
1973, revised ed. 1988; idem, The architecture of 
power: palaces, citadels and fortifications, in Architecture of 
the Islamic world, ed. G. Michell, London 1978, 48- 
79; R. Hillenbrand, 'La dolce vita' in early Islamic 
Syria. The evidence of later Umayyad palaces, in Art 
History, v (1982), 1-35; J. Scarce, The royal palaces of 
the Qajar dynasty: a survey, in Qajar Iran, ed. E. 
Bosworth and C. Hillenbrand, Edinburgh 1983, 
329-51; R. Ettinghausen and Grabar, The art and ar- 
chitecture of Islam 650-1250, Harmondsworth 1987 
(pictures of many early Islamic sites); Ars Orientalis, 
xxiii (1993) (special issue on pre-modern Islamic 
palaces, with articles on many of the sites and 
periods mentioned above); R. Hillenbrand, Islamic 
architecture, Edinburgh 1994 (esp. ch. VII, palace]; 
S.S. Blair and J. M. Bloom, The art and architecture of 
Islam 1250-1800, London and New Haven 1994 
(colour plates of many palaces from the later 
period); Bloom, Palace, Islamic lands, in Dictionary of 
art, in press. (Sheila S. Blair) 

SARBADARIDS, the name given to the rulers of 
the Bayhak (Sabzawar [q.v.]) region of 
Khurasan and a section of their followers, in the half 
century between the death of Abu Sa'Td [q.v.] and the 
conquest of Persia by Timur [q.v.]. (On the name, 
often also found as Sarbadal ("refactory"), see e.g. 
Nawa'i, note in Samarkand!, 432.) The Sarbadarid 
regime has been variously viewed as a robber state, a 
social revolutionary movement animated by a strong 
Mahdist impulse, and a type of Shi'r "republic" (see 
Roemer, 38-9). It can most usefully be seen as an at- 
tempt at self-government among the indigenous 
population of western Khurasan, faced with the 
disintegration of Mongol rule. 

The Sarbadarid uprising started in Bashtln, in the 
district of Sabzawar, on 9 Sha'ban 737/13 March 
1337 (FaryumadI, 347; cf. Samarkand!, 147). c Abd 
al-Razzak, son of a respected local notable, murdered 
a tax official and, in an effort to forestall the conse- 
quences, rose in revolt. The readiness of others to 
follow c Abd al-Razzak, who is painted as a rash and 
abusive character, and later his more sophisticated 
brother, Wadjih al-DIn Mas c ud, can and has been ex- 
plained by the financial oppression oj the wazir of 
Khurasan, c Ala> al-Din Muhammad (Amuli, 181-2). 
'Ala 5 al-Din's exactions were driven by the need to 
finance the aspirations of the Cingizid pretender 
Tughay Timur, whose vain efforts to impose himself 
as heir to the Ilkhanate depleted Khurasan's resources 
and caused the Mongol leaders to neglect the Sar- 
badarid uprising until it was too late. 

c Abd al-Razzak was himself devoid of any great 
political purpose, but was able to gain control of Sab- 
zawar (12 Safar 738/9 September 1337), where 
several members of 'Ala 3 al-DIn's family were killed 
(FaryumadI, 347, 326). 'Abd al-Razzak's immoral 
behaviour led to his death late in 738/June 1338 at the 
hands of his brother Mas'ud, who represented the 
wider ambitions of the rural landowners of the region, 
anxious to throw off Mongol rule. The growing band 
of soldiers, young men and '■ayyars [q.v.] who eagerly 
joined him were rewarded by an egalitarian distribu- 
tion of the spoils gained from a series of raids, directed 
particularly against the Dja'un-I Kurban, which 
culminated in the capture of Nishapur (Amuli, 182). 
More significantly, Mas'ud sought to strengthen 


the rather flimsy bases of his authority, and also to at- 
tract further support, by involving Shaykh Hasan-i 
Djurl, a disciple of the MazandaranI Shl'T darwish, 
Shaykh Khalifa, in his cause. Although there seems 
little doubt that Djurl had not become master of a 
radical underground movement, he recognised that 
the grievances of his adherents and of the oppressed 
Muslims of Khurasan in general could best be re- 
presented if he associated himself with the protests 
being voiced by the outlaws of Sabzawar. Thus began 
the uneasy alliance between the Sarbadarids and the 
Shaykhis that gives the regime its distinctive character 
(Aubin, in Stir, v/2 [1976]). 

Tughay Timur, in whose name coins had con- 
tinued to be minted, even in Sabzawar (Masson 
Smith, 109, 201), could no longer afford to ignore the 
Sarbadarids, and after an exchange of messages that 
reflected the new religious dimension contributed by 
Hasan-i DjurT (Mirkh w and, 614), the two sides met 
near the Gurgan river. Tughay TTmur's army was 
routed. 'Ala 3 al-DIn Muhammad, his wazit, was pur- 
sued and killed in Kabud Djama on 23 Sha'ban 742/1 
February 1342 (Ibn Yamln, 569). The Sarbadarids 
now coined in the name of Tughay TTmur's rival 
Sulayman, the protege of the Cubanids [?.».], and 
sought to expand their domains further at the expense 
of the Karts [g.v.] of Harat. This evidently coincided 
with a crisis in the relationship between Mas'ud and 
the Shaykhis, for during a battle near Zawa on 13 
Safar 743/18 July 1342 (Ibn Yamln, 570), Hasan-i 
Djurl was killed, allegedly on Mas'ud's orders 
(Faryumadi, 348), turning victory into defeat. 
Undeterred, Mas'ud soon launched an invasion of 
Rustamdar and Mazandaran, but the campaign end- 
ed in disaster, and he was killed in Dhu'I-Ka'da 
743/April 1343 (Amull, 183-9; Faryumadi, 348). The 
following year, coins in Tughay TTmur's name were 
again minted in Damghan, which had previously 

The leadership was now assumed by one of 
Mas'ud's commanders, Muhammad Aytlmur, who 
was obliged to acknowledge once more the sovereign- 
ty of Tughay Timur. Although he had shown himself 
a competent ruler, he was murdered at the instigation 
of the darwish,, in Muharram 747/April 1346 
(Faryumadi, 348), who perhaps favoured a more ag- 
gressive policy. This inaugurated a period of instabili- 
ty as the Sarbadars, broadly loyal to the family of 
Mas'ud Bashtlni, and defending the interests of the 
Sarbadarid troops, vied for power with the darwishs, 
whose chief spokesman (though not their spiritual 
leader), Kh w adja Tadj al-DIn 'All b. Shams ai-DIn 
Cishumi , finally agreed to become ruler ( 1 6 Sha'ban 
748/21 November 1347; Faryumadi, 348). Tadj al- 
DIn 'All brought a measure of order to Sarbadarid af- 
fairs. He issued the first independent Sarbadarid 
coinage (748/1348), SunnI in type, and followed a 
programme of economic and moral reform designed 
to appeal to the wide range of opinion contained with- 
in the Sarbadarid state (Masson Smith, 130-2). In 
general, he was also able to reach agreements with his 
various neighbours. In 752/1351, following the incor- 
poration of Harat into the Caghatayid sphere of influ- 
ence, 'All deemed it prudent to mint coins once more 
in the name of Tughay Timur. This appeasement of 
the Mongols, in top of his own autocratic methods, 
led to his murder on 28 Shawwal 752/18 December 
1351 (Ibn Yamln, 571). 

One of the assassins, Yahya KarrabI, a landed pro- 
prietor of the district, resumed military hostilities and 
after several confrontations with Tughay Timur, 
whose ordu was weakened by the Black Death, he 

feigned a willingness to make peace and treacherously 
murdered the last of the Cingizids in his camp at Pul-i 
HadjdjI Khatun, near Sultan Duwin, on 16 Dhu'I- 
Ka'da 754/13 December 1353 (Dawlatshah, 237-8). 
KarrabI later came to terms with the Mongol rulers of 
eastern Khurasan, in a coalition against the malik of 
Harat, but was almost immediately murdered, on 10 
Muharram 759/23 December 1357 (Ibn Yamln, 568; 
Aubin, in Turcica, viii [1976], 39-41). 

Another period of instability followed, again mark- 
ed by a series of coups and political murders, the 
precise chronology of which remains elusive; even the 
most reliable sources are inconsistent and mutually in- 
compatible. This may partly be due to the difficulty 
(even for contemporaries) of distinguishing the 
nominal from the actual periods of power enjoyed by 
the different leaders, the two most prominent of 
whom, Haydar-i Kassab and Hasan-i Damghanl. 
both installed their own candidates before seizing 
power. One of these, Lutf Allah son of Wadjlh al-DIn 
Mas'ud, was the last member of this family to rule. 
Internal divisions spawned external threats. Amir 
Wall, son of the former governor of Astarabad, took 
advantage of the situation to extend his control in the 
area, ostensibly on behalf of Tughay TImur's son 
Lukman. He defeated the forces sent by Haydar-i 
Kassab and was able to issue coins in his own name 
in Astarabad in 759/1358 (Morton, 257). A large 
force led by Hasan-i Damghanl was also defeated 
(Faryumadi, 330). Damghanl faced further dif- 
ficulties. Darwish 'Aziz, a follower of Hasan-i DjOri, 
installed himself in Mashhad, where his devotions at- 
tracted followers. With their support, he rose in revolt 
and seized Tus. Contrary to some assertions, there is 
no numismatic evidence that he tried to establish a 
Mahdist state there (Morton, 257). He was expelled 
from Khurasan, but Kh w adja 'All b. Mu'ayyad, son 
of a Sabzawarl notable, rose in revolt in Damghan, 
sought Darwish 'Aziz's help and together they cap- 
tured Sabzawar. Hasan-i Damghanl was murdered, 
evidently in 763/1362. 

With the accession of Khwadja 'All, the Shi'ism of 
the Sabzawaris first found explicit political expres- 
sion. He announced his rule with a new issue of coins 
bearing a Shi*! formula, and instituted the practice of 
leading a horse out twice daily for the Sahib-i 
Zaman's expected arrival. Towards the end of his 
reign he invited an ImamI scholar, Muhammad b. 
Makkl al-'Amill [?.».], to Sabzawar to preside over 
the establishment of Twelver Shi'ism in the Sar- 
badarid realm, anticipating a similar move by the 
Safawids [q.v.\ a century and a half later. Ibn Makkl 

fikh to guide Khwadja 'All (Mazzaoui, 66-7). The 
earlier measures probably reveal the influence of Dar- 
wish 'Aziz and the Shaykhiyya, still hoping for a new 
dispensation; nevertheless, Kh w adja 'All's commit- 
ment to this view was lukewarm, and within a year he 
had engineered the removal of Darwish 'All (18 Rabr c 
I 764/5 January 1363; Faslh, 95-6) and driven out the 
Shaykhiyya. even going so far as to desecrate the 
tombs of Shaykh Khalifa and Hasan-i Djurf 
(MIrkh"and, 624). The flight of Darwish 'All's suc- 
cessor, Rukn al-DIn, to the court of Shah-i SJhudja' 
the Muzaffarid [q.v.] led to an invasion of Sarbadarid 
territory with Muzaffarid support, in which Sabzawar 
was captured in 778/1376. Kh w adja 'All was obliged 
to seek refuge with Amir Wall, who helped him regain 
the city on 8 Radjab 781/20 October 1379 
(Faryumadi, 331-3); but Kh w adja 'All was left with a 
greatly reduced realm. 

Two years later, in 783/1381, Kh w adja 'All took 


the opportunity of preserving himself by entering 
TImur's service. On his death in Huwayza in 
788/1386, the Sarbadarid realms were divided 
amongst several leaders, who also served TTmur, 
particularly in the administration of Lower c Irak. In 
808/1405, an attempt by a relative of Kh w adja C AU to 
claim his "hereditary rights" to the former Sar- 
badarid territories ended in his execution and the sack 
of Sabzawar. Another member of the family later won 
fame as the poet Amir Shahl, 

The Sarbadarid regime channelled a number of 
currents flowing through Khurasanian society in the 
8th/ 14th century. Its history reveals something of the 
aspirations of the rural population and their leaders, 
the activities of the futuwwa [q.v.] organisations, the 
growing role of Sufi shaykhs in political life, and the 
hesitant emergence of ShlSsm as an expression of 
local particularism. Many of these themes are ap- 
parent in the poetry of the panegyrist Ibn Yamin 
[q.v.], who also praises the building works of rulers 
such as Tadj al-Dln C A1I and Yahya KarrabI (Ibn 
Yamin, 30, 86). The latter's restoration of the kandts 
[q.v.] in Tus (Dawlatshah, 282-3) is symptomatic of 
the Sarbadarids' vain but heroic attempt to counter 
the negative effects of Mongol rule in Khurasan. 

Bibliography: 1. The main sources (in approx- 
imately chronological order) are Ibn Battuta, 
Travels, ed. and tr. Sir Hamilton Gibb, iii, Cam- 
bridge 1971, 574-7; Amull, TaMkh-i Ruydn, ed. M. 
Sutuda, Tehran 1348/1969; Ibn Yamin, Di'wdn, ed. 
Husayn- C AM Bastani Rad, Tehran 1344/1965; 
Faryumadl, Dhayl-i madjma^ al-ansdb [by 
Shabankara'I], ed. Mir Hashim Muhaddith, 
Tehran 1363/1984; Hafiz-i Abru, Madjmu'a, partial 
ed. F. Tauer, Cinq opuscules de Hafiz-i Abru, Prague 
1959; Faslh-i Kh w afl, Mudpnal-i Fasihi, iii, ed. M. 
Farrukh, Mashhad 1341/1962; Samarkandl, Matla c 
al-sa c dayn, ed. c Abd al-Husayn Nawa'I, Tehran 
1353/1974; Dawlatshah; Mlrkh^and, Rawdat al- 
safd\ v, ed. Tehran 1339/1960, esp. 600-25; 
Kh w andamlr, Habtb al-siyar, iii, ed. M. Dabir- 
Siyakl, Tehran 1362/1984, esp. 356-67. 

2. Studies. For the earlier work, see V.F. 
Buchner, EI 1 serbedars. Two full-length 
monographs have since appeared, namely I. P. 
Petrushevskii, Dvizhenie serbedarov v Khorasane, in 
Ucenye Zapiski Instituta Vostokovedeniya Akademii Nauk 
SSR, xiv (1956), 91-162, tr. Karlm Kishawarz, 
Nahdat-i Sarbaddran dar Khurasan, in Farhang-i Irdn- 
zamin, x (1341/1962), 124-224, and J. Masson 
Smith Jr., The history of the Sarbaddr dynasty 1336- 
1381 A.D. and its sources, The Hague and Paris 
1970. The latter attracted useful comments from H. 
Arroyo, Complement to the numismatic history of the Sar- 
baddr dynasty, in Seaby's Coin and Medal Bulletin 
(1975), 302-4, and A.H. Morton, The history of the 
Sarbaddrs in the light of new numismatic evidence, in NC, 
7th ser., xvi (1976), 255-8. Knowledge of the period 
has been greatly advanced by a number of studies 
by J. Aubin, notably, La fin de I'Etat sarbaddr du 
Khorassan, in JA (1974), 95-118; idem, Aux origines 
d'un movement populaire medieval: le Cheykhisme du 
Bayhaq et du Nichdpour, in Stir, v/2 (1976), 213-24; 
idem, Le khanat de Cagatai et le Khorassan, 1334-1380, 
in Turcica, viii (1976), 16-60; idem, art. "Abd-al- 
Razzdq Bdsttni, in EIr, i (1985), 153-4; and idem, Le 
quriltai de Sultan-Mayddn (1336), mJA (1991), 175- 
97. The most recent survey is by H.R. Roemer, The 
Jalayirids, Muzaffarids and Sarbaddrs, in Camb. hist. 
Iran, vi (1986), esp. 24-39, 49-50. c Abd al-RafF 
Haklkat, Ta^nkh-i djanbish-i Sarbaddran wa dlgar 
djanbish-hd-yi Irdniydn dar karn-i hashtum-i hidjri, 

2 Tehi 


1363/1984, is unaware of Masson 
id Aubin's work. See also Rashid YasimI, /. 
n-i Yamin, Tehran 1303/1924; M.M. Maz 
of the Safawids: frism, $ufism, c 
' ■— (C.P. Mel 


SARDAB (p.), literally "cool water", often found 
in the Arabised form sirddb, an underground 
chamber used for keeping cool during the ex- 
treme heat of e.g. the c IrakI or Persian summers. 

Such building constructions are an ancient feature 
of Middle Eastern life, being found amongst the 
Egyptians of Pharaonic times and in Babylonia. Ex- 
amples of them have been found in the remains of the 
early c Abbasid palace at al-Ukhavdir [q.v.] and at al- 
Mu c tasim's palace, the Djawsak al-Khakanl, at 
Samarra. At Baghdad until recent times, traditional- 
type houses had a semi-basement vaulted cellar sunk 
into the ground some 1.5 m/5 ft. below the surface of 
the ground. This was ventilated by a shaft on the 
north side of the chamber running up to the highest 
point of the house in order to catch the cooler north 
wind (this shaft being called in mediaeval Arabic 
badahandi or bddandj., Arabised from Persian bddhandj_ 
[see badcir in Suppl.], and also by several small win- 
dows just above ground level. People would spend the 
hottest hours of the day, from mid-morning till 
sunset, in these rooms. In the equally hot and humid 
coastlands of Khuzistan, Fars and Kirman in 
southern Persia, such chambers are also found and 
are called zir-i zamin ("subterranean"); again, they 
are usually provided with a ventilating shaft or bddgir. 
It was more elaborate forms of these which G.N. Cur- 
zon encountered at Shushtar on Khuzistan. under the 
name of shabedan or shevedan. These were hewn into the 
bedrock on which the town stood to depth of 18 m/60 
ft. or more, reached by steps, and with a ventilating 
shaft and with light coming from a circular light in the 
vaulted roof; duringjuly and August, the inhabitants 
lived almost entirely in these subterranean chambers 
(Persia and the Persian question, London 1892, ii, 367). 

sirddb ( 


id of ui 

ir passage. 

In Ibn Battuta, Rihla, i, 264-5, tr. Gibb, i _ 
dab is used for the passage constructed by c A 5 isha and 
leading from the Prophet's mosque in Medina to what 
was the house of Abu Bakr outside the mosque 
precincts (cf. the khawkha or private entrance to the 
mosque, regarded as a special privilege (khasisa) of 
Abu Bakr's; see Wensinck, A handbook of Muhammadan 
tradition, 6, and C.E. Bosworth, Al-MaqrizT's "Book of 
contention and strife concerning the relations between the Band 
Umayya and the Band Hashim", Manchester n.d. 
[1981], 86-7, 139). 

Bibliography (in addition to references given in 
the article): C. Niebuhr, Voyage en Arabic et en d'autres 
pays circonvoisins, Amsterdam 1780, ii, 279; Sir R. 
Ker Porter, Travels in Georgia, Persia, Armenia, An- 
cient Babylon..., London 1821-2, ii, 261; J. S. Buck- 
ingham, Travels in Mesopotamia..., London 1827, ii, 
192, 210; Dozy, Supplement, i, 647; A. Badawy, Ar- 
chitectural provision against heat in the Orient, \nJNES, 
xvii (1958), 127. See also khaysh; mirwaha; 
badgIr in Suppl. (C.E. Bosworth) 

SARDANIYA, the most usual Arabic transcription 
of the place name Sardinia, the second largest in 
size of the island in the western Mediterra- 
nean. The author of the EI 1 article mentions also the 
transcription Sardaniya. 

Amongst the reasons impelling the Arabs, who had 
just invaded North Africa at the beginning of the 7th 
century A.D., to conquer Sardinia, was the fact that 
it was an appendage of the Byzantine Exarchate of 


Africa. To this first reason, one might add a desire to 
acquire the silver mines required for a bimetallic 
monetary system, one based on both gold and silver, 
and timber for constructing ships. In Sardinia, there 
were silver mines in the district of Sulcis, not far from 
Cagliari, whilst the whole island was practically 
covered with forests. 

The historical information. 

At least eleven different Arabic sources describe, in 
a summary fashion and without details, a series of in- 
cursions stretching from the 8th to the 12th century 
A.D. and taking place at the following dates: 84/703- 
4; 87/705-6; 89/707-8; 92/710-11; 114/732; 117/735; 
119/737; 135/752-3; 201/816-17; 206/821-2; 322/933- 
4; 323/934-5; 405/1014; 406/1015; and 446/1054. The 
sources in question are 1. the K. al-Imdma of Ps.-Ibn 
Kutayba, attributed to Ibn al-Kutiyya (d. 367/977); 
2. al-DabbI(d. 599/1203); 3. Yakut (d. 626/1233); 4. 
Ibn al-Athir (d. 630/1233); 5. Abu '1-Fida 5 (d. 
732/1331); 6. al-Nuwayri (d. 732/1332); 7. Ibn 
c IdharI(d. 750/1349); 8. al-Dhahabl (d. 748/1348); 9. 
Ibn Khaldun (d. 808/1406); 10. al-Makrizi (d. 
845/1442); and 11. Abu '1-Mahasin Ibn Taghribirdl 
(d. 874/1470). 

On the Sardinian side, the sources provide no 
details at all on an Arab presence in the island except 
for general statements that the Sardinians always 
fought with great courage against the invaders. One 
fact only is reported in detail by sources worthy of 
credence, such as the Venerable Bede and Paul the 
Deacon, author of the Historia Langobardorum. This is 
the transfer of the body of St. Augustine from Cagliari 
to Pavia, in northern Italy, at the intervention of the 
King of the Lombards, Liutprand, who between 721 
and 726 purchased the relics of the saint so as not to 
leave them in the hands of the Saracens. 

Epigraphy. In the Cagliari Archaeological Museum 
there are three Kufic inscriptions. The first is a 
funerary stele of the prismatic type called mkabriyya in 
the Ma gh rib, which comes from Assemini, a small 
town to the north of Cagliari. On it there is the name 
of a certain Maryam, daughter of c Atiyya al-Sarradj, 
deceased in 470/1077. The second was found at 
Cagliari, in the neighbourhood of the palace which 
was formerly that of the Viceroy; it is a fragment, 
whose date may be identified with the former one. 
The third stems from Olbia, in the north. It is a large 
rectangular slab, with the top part and the right-hand 
side missing. The name which can be partly read is 
Mustafa Muhammad al-M . . . These inscriptions are 
carved on a kind of limestone which is found in Sar- 
dinia, but the dates fall after the period of an Arab 
presence, which ended towards 1070. 

Coins. Given the small number of pieces of money 
(19 in all) kept in the Archaeological Museum at 
Cagliari, one is tempted not to take them into ac- 
count. But 11 of the coins belong to the period be- 
tween 700 and 778; moreover, 7 are copper, a metal 
which was not transported about and which no-one 
hoarded in a monetary form. 

Toponyms. Numerous scholars have categorically 
stated that Sardinian toponomastic has no Arabic 
terms at all. There are nevertheless some elements 
which invite reflection. The etymology of Alghero 
could be "grotto, cave" in Arabic. There exists in fact 
in the outskirts of this town the "grotta di Nettuno" 
to which one descends by means of "la escala del 
Cabirol" of 600 paces and which has been known 
since Antiquity for its internal lake and its enormous 
halls with fantastic rock-shapes. There are at least two 
other examples of place names involving numerals: 

and also 

nto a context of place nan 
:o ("the 5th") and ends w 

'hich could n 
which is found inserted ir 
which begins with Quin 
Decimo ("the 10th"). 

In brief, the dates given, with not exactitude, by the 
Arabic historians point rather to raids, except perhaps 
for a half-century of occupation in the area around 
Cagliari, between 700 and 750 A.D. The surviving 
traces — epigraphic, numismatic and some place 
names — have dates varying between the 8th and the 
1 lth century, which could suggest the existence of col- 
onies which are not, unfortunately, confirmed by 

Bibliography: G. Oman, Vestiges arabes en Sar- 
daigne, in ROMM (1970), 175-84; idem, Iscrizioni 
arabe di Sardegna, in Atti della Settimana di Studi 
Mediterranei Medioevali e Moderni, Cagliari 1979, 213- 
27; idem, Monele con iscrizioni arabe nel Museo Ar- 
cheologico Nazionale di Cagliari, in Annali dell'Istituto 
Italiano di Numismatka, xv (1968), 115-17; Maria 
Giovanna Stasolla, Arabi e Sardegna nella storiografia 
araba del Medioevo, in Studi Magrebini, Naples, xiv 
(1982), 1-40, Appendice II, see Bibl.; Emanuela 
Boiardi, Gli Arabi in Sardegna nel secolo VIII e la trasla- 
zione del corpo di S. Agostino, diss. Bologna 1978-9, 
unpubl. (G. Oman) 

SARDAR (p.), often Arabised as Sirdar, 
"supreme military commander", literally 
"holding or possessing the head", i.e. chief or leader. 
It was borrowed in the military sense by the Turks, 
who, however, sometimes derive it in error from sirr- 
dar ("the keeper of a secret"). Through Turkish it has 
reached Arabic, and in a letter written in 989/1581 by 
"one of the princes of the Arabs (of Yaman)" occurs 
the phrase wa- c ayyana sardar"" c ala 'l-'-asakir ("and he 
appointed a commander over the troops"), on which 
Rutgers comments ' ' Vocabulum sardar, quod Persicae 
originis est, dusem exercitus significat". The abstract 
substantive sardariyyat in the sense of the post or office 
of commander of an army also occurs; and it was 
doubtless owing to the familiarity of the Arabic- 
speaking people of Egypt with the borrowed word that 
it was selected as the official title of the British 
commander-in-chief of the Egyptian and Sudanese ar- 
mies in the periods of the Protectorate and the Con- 
dominium. In Persia the word was until the early 20th 
century much used as a component part of honorific 
titles, such as Sarddr-i Zafar and Sarddr-i D^ang. In 
British India it was used generally of the (Indian) 
commissioned officers of the army as a class. Sardar log 
meant "the (Indian) officers of a corps or regiment". 
It was formerly applied to the head of a set of 
palanquin-bearers, and it was still in the early 20th 
century applied to the valet or body-servant of a Euro- 
pean in northern India, as the chief of his household 
servants. Sardar Bahadur was a title of honour attached 
to the first class of the Order of British India, an order 
confined to Indian commissioned officers of the army. 
Bibliography: The standard lexica of Persian, 
Turkish and Urdu; R. Dozy, Supplement aux diction- 
naires arabes; A. Rutgers, Historia Jemanae sub Hasano 
Pascha, Leiden 1838; H. Yule and A.C. Burnell, 
Hobson-Yobson, 2nd edition by W. Crooke, London 
1903, 841. _ (T.W. Haig') 

SARDHANA, a town, also the centre oiatahsil, in 
the Meerut [see mirath] District of northwestern In- 
dia, now in the Uttar Pradesh State of the Indian 
Union. The town is situated in lat. 29° 09' N., long. 
77° 36' E. and lies some 19 km/12 miles to the north- 
west of Meerut town. 


It achieved fame in the later 18th century, when 
Walter Reinhardt, called Sombre or Samru, of 
Luxemburg origin, after having been a mercenary in 
both French and British service, received from MTrza 
Nadjaf Khan, general of the Mughal Emperor Shah 
c Alam II [q.v.], the pargana [q.v.] of Sardhana [q.v.]. 
This became, after his death in 1778, the centre of a 
small, virtually independent principality, kept up by 
his remarkable Indian wife, the Begam Samru [q.v.], 
surviving within British territory as a distinct entity 
and family estate until her death in 1836, when it was 
resumed, eventually being granted to Djan Fishan 
Khan, formerly leader of the Sayyids of Paghman in 
Afghanistan and his family. In 1961, Sardhana had a 
population of 16,563, and the tahsil (in whose rural 
areas the Muslims are especially represented) one of 

Bibliography: Imperial gazetteer of India 1 , xxii, 

104-7; P. Spear, Twilight of the Mughuls, studies in late 

MughulDelhi, Cambridge 1951, 115, 143, 152. Uttar 

Pradesh District Gazetteers, Meerul, Allahabad 1968, 

44 ff., 48-9. See also samru, begam. 

(C.E. Bosworth) 

SARDJ (a.), a masc. noun (pi. surudj) denoting the 
horse saddle, and this uniquely; from the same root 
s-r-dj_, there is saridja for a mule or camel saddle. From 
this root stem also the verbs saradja, also forms II and 
IV, for "to saddle amount" (also used are kasd al-sardj 
and, for unsaddling, ramd al-sardj), and then sarrddj 
and surudji for the saddle-maker and seller of saddles, 
and sirddja and surudjiyya for the craft of making sad- 
dles. A horse which is saddled is musarradj_. Every town 
of the Arab lands had its own quarter or market for 
saddle-makers (suk al-sarrddjiri), usually located near 
the ways out of the urban area in order to facilitate 
traffic with the countryside. 

Amongst the Arabs, the traditional saddle is made 
up of the saddle-bow ( c azam al-sardj), a basic frame 
made out of two wooden curved pieces connected 
together by their ends and forming a vaulted shape on 
the mount's back. The interior of this vaulted shape 
is padded, and on this base is placed the leather seat 
(kursi) of the saddle. The front of the pommel (karbds, 
pi. kardbis, and vars. karbus, karbui) is slightly raised 
and the reins can be fastened. to it. The back pommel 
(karbus mu^akhkhar) supporting the rider's reins is 
generally very high. Each lateral facing of leather or 
quarter, on which the backside of the rider is set 
down, receives the stirrup-leather (sayr al-rikdb) bear- 
ing the stirrups and the stirrup-holder (ribdt, pi. rubut). 
Between the mount's back and the saddle is placed the 
saddle-cloth (mirshaha), with wool preferred to cotton, 
thus avoiding rubbing, callosities and wounds. The 
whole of the saddle is then fixed on the animal's back 
by the saddle-girth (mihzam, wadin, c adiala), and a long 
chest tether (labab) or breast-strap keeps it from mov- 
ing back. At the opening of the 8th/14th century, the 
author of a work on hunting, Ibn Mangll (see the tr. 
ofF. Vire, Delachasse, Paris 1984, 43-4), categorically 
forbids use of the high back pommel of the saddle, 
which the Persians, he alleges, supposedly introduced, 
because it impedes the drawing of a bow, both in 
hunting and in combat; he denounces its nine major 
faults and adjures the Mamluks to reject this type of 

The saddle could be decorated with gleaming or- 
naments and pieces of embroidery and copper stud- 
could be covered 
broidered with gold thread. 

The saddle strapped 
the basis for figurative 


m (karabasun 

"your saddle has slipped off" for "your affairs are go- 
ing badly"; ram fi 'l-sardl "see me firmly in the sad- 
dle", meaning "I have succeeded"; and, the con- 
trary, rani ward 7 al-sardj "see me behind the saddle", 
meaning "I have failed/lost". 

Bibliography: See the exhaustive bibls. to faras 
and furus.vva. (F. Vire) 

SAREKAT ISLAM, a Muslim movement in 
the Netherlands East Indies which flourished 

The establishment in 1912 of Sarekat Islam opened a 
new era for both Islam and political mobilisation in 
the Dutch East Indies. It actually grew out of an 
association with more limited aims, the Sarekat Dagang 
Islam ("Association of Islamic Traders"), set up in 
1909 by Raden Mas Tirtoadisoerjo, a Javanese 
aristocrat and merchant whose trading company was 
then being liquidated. He and other Javanese mer- 
chants set this up as a co-operative trading association 
to counter Chinese economic dominance; from the 
late 19th century, the Chinese had begun to take over 
even those small industries (such as the production of 
batik cloth and kretek cigarettes) which had till then 
been Javanese-dominated. The association organised 
successful anti-Chinese boycotts and propaganda, 
leading to government action against it. One of its 
members, the batik manufacturer Hadji Samanhoedi, 
consequently turned to Hadji Omar Said Tjokro- 
aminoto to rebuild the association. Tjokroaminoto, 
born in 1882, was the son of a relatively minor official 
in the colonial Javanese bureaucracy, and had himself 
been trained at the training school for native officials 
(OSVIA, Opleidingsschool van Inlandschen Amblenaren). 
However, he spent only three years in the 
bureaucracy before moving on to other occupations 
which led him to travel widely across Java. He now es- 
tablished a re-formed organisation, called Sarekat 
10 September 1912. The original comi 


; derm 

Oetoesan Hindia ("The Indies Messenger"). But Sarekat 
Islam soon became a mass movement whose member- 
ship went far beyond the elite group responsible for its 
foundation. It grew phenomenally and drew in 
diverse elements: not only the small group of Muslim 
entrepreneurs from whom the founders had been 
drawn, but also Muslims from the world of the 
mosque school, Islamic reformists, and, increasingly, 
the peasant masses. Already by 1914 it had over 
360,000 members, and by the time of its first national 
congress in June 1916 it had recruited more than 
80,000 members outside Java. 

Islam had had a leading role in large-scale political 
mobilisation for many centuries. Yet this mobilisation 
had been pre-modern in its organisational form, being 
led by traditional elites such as the hereditary 
aristocracies of the Indies or the elites associated with 
mosque schools and tarekats. In Java, mobilisation in 
the name of Islam had been led by princes, and Islam 
had been largely subordinated to pre-Islamic Javanese 
political values. Many Islamic concepts had been 
redefined to accommodate to a highly monarcho- 
centric policy in which service to the ruler (ngawula) 
was the supreme moral virtue, and Javanese rulers 
claimed to be endowed with both wahy [q. v. ] and the. 
Light of Prophecy. From the foundation of Sarekat 
Islam it is clear that Islam had freed itself from the old 
royalist ideology and patrimonial forms of mobilisa- 
tion, and had done much to make possible an in- 
digenous political life based on associational forms, on 
Gesellschaft rather than Gemeinschaft. Indonesia's links 
with the heartland of Islam had also greatly 


strengthened since the late 19th century through a 
large increase in those making the Pilgrimage, who 
returned with a stronger commitment to Islamic 
norms and also brought back a better knowledge of 
political and military developments overseas. At the 
local level, the modernising aspect of Sarekat Islam was 
less evident, indeed was sometimes replaced by a pro- 
test against the cost to the peasantry of economic 
modernisation under the colonial aegis, or against the 
Chinese money-lenders to whom indebted or tax- 
burdened peasants had to turn. The years 1913-14 
saw a peak of anti-Chinese violence. To say "I am a 
Muslim" at this period was a way for those at the bot- 
tom of the colonial racial hierarchy (below Dutch, 
Eurasians, and Chinese) to claim a more positive 
identity than that allocated to them by the Dutch term 
inlander (native). Sarekat Islam by its very name tapped 
this sense of identity and grievance. In addition, 
Tjokroaminoto was a charismatic leader who became 
a focus for the messianic beliefs that had been a prom- 
inent feature of traditional movements, whether 
Islamic or Javanist. By now the organisation, rather 
ambitiously, claimed a membership of two millions. 

The role of Islam in introducing more modern con- 
cepts into political mobilisation in the Indies was cer- 
tainly not uncontested. Socialist ideas first developed 
a significant presence in the Indies just at the time 
Sarekat Islam was established. The Dutch Socialist 
Sneevliet founded the ISDV (Indische Sociaal- 
Demokratische Vereeniging, Indies Social Democratic 
Union) in May 1914 and, after a lack of success with 
the Eurasian community, turned his attention to 
Sarekat Islam. From 1916, the Semarang branch of 
Sarekat Islam was the centre of the Socialist wing, led 
by the 17-year-old Semaun. Semarang was then the 
most progressive and liberal city in the Indies, the 
centre of an embryonic proletariat (associated with the 
railways and service industries) and of trade union ac- 
tivity. At the second Sarekat Islam conference in 1917, 
the Semarang group moved to insert into the 
organisation's programme the combating of "sinful" 
(i.e. foreign) capitalism, later to be extended to 
capitalism in general. The position of the Semarang 
group was strengthened after the October Revolution 
in Russia (which led to a dramatic rise in membership 
of Sarekat Islam), and they were joined by two central 
Javanese aristocrats, Darsono and Surjopranoto. An 
opposing group under Abdul Muis and Hadji Agus 
Salim (who had originally joined Sarekat Islam as a 
government spy but been converted to its cause) were 
branded as tools of the colonial government. At the 
third Sarekat Islam conference, Tjokroaminoto suc- 
ceeded in reconciling the two factions. 

1919 was a turbulent year on Java, with sugar plan- 
tations set on fire and peasants refusing to perform 
forced labour. In mid-year there occurred local in- 
cidents first in Sulawesi and then in West Java. The 
latter case ended in the slaughter by government 
troops of a Garut landowner, Hadji Hasan, and 
members of his family, after they had shut themselves 
into their house and refused to hand over the rice tax. 
It was claimed that there was within Sarekat Islam a 
clandestine "Section B", dedicated to the overthrow 
of colonial rule. As planters and others in the Euro- 
pean community panicked at the spectre of a Muslim 
conspiracy, the government reacted strongly, leading 
many of the middle-class Muslims who had founded 
Sarekat Islam to leave the organisation. The relatively. 
liberal attitude to Islam initiated by Snouck Hurgron- 
je's policy advice to the colonial government was 
under serious challenge. 

In May 1920 the ISDV changed its name to 

Perserikatan Komunis di Inida ("Indies Communist 
Union", later "Party") and in July 1920 the Com- 
intern passed a resolution that pan-Islamic move- 
ments be opposed as strengthening Turkish im- 
perialism, despite Sneevliet's plea that cooperation 
continue. Sarekat Islam had already set up a committee 
in support of the caliphate and it was becoming clear 
that the division between the Socialist and pan-Islamic 
wings of the party was sharpening. Tjokroaminoto, 
the chief architect of compromise, was arrested on the 
grounds of supposed involvement in the Garut affair. 
During his absence, on the occasion of the October 
1921 conference, Hadji Agus Salim introduced party 
discipline in Sarekat Islam, leading to a split between 
the Islamic and Socialist-Marxist movements and to a 
subsequent battle between them for control of the 
branches. Despite the organisational break, at the 
popular level Islamic communist movements con- 
tinued to exist, as for instance that led by the extreme- 
ly popular Haji Misbach in Solo, whose peasant 
following were attracted by the idea of reinstituting a 
golden age of justice for all. When Sarekat Islam impos- 
ed party discipline, he opted for the PKI, and was 
subsequently banished by the government. 

In January 1922 a strike of employees in the 
government-run pawnshops, the first large-scale 
union-sponsored work stoppage, took place. It was 
supported by a number of political organisations, and 
the subsequent government crack-down fell most 
heavily on Sarekat Islam in the arrest and deportation 
of its leadership. At this juncture the gulf between the 
aspirations of the Indonesian oppositional movements 
and the limits set by the colonial government was 
harshly illuminated. Semaun subsequently resumed 
leadership of the decapitated PKI, and Tjokro- 
aminoto that of Sarekat Islam. The latter now published 
his book Islam and Socialism (Islam dan Sosialisme) which 
represents a hardening of his attitude to Marxism. In 
it he attempts to demonstrate that Islamic socialism is 
the most perfect kind, not only by reference to 
Muhammad's teachings but also by a reconstruction 
of society under the Orthodox Caliphs. He says that 
the state owned land, which was the same as owner- 
ship by the people; that society was neither autocratic 
nor bureaucratic; and that the ; 




s in fore. 

prohibition ofribd, the institution of zakat, injunctions 
to charity, and the egalitarianism evident in the com- 
munal prayers and during the Pilgrimage. In this 
book, Tjokroaminoto strongly attacks historical 
materialism as denying God, and deifying material 
things. He condemns Bebel's famous dictum to the ef- 
fect that it is not that God created man, but that man 
has invented God. Much of the work is organised not 
around specifically socialist ideas but around the 
ideals of Liberty, Equality, and Fraternity, indicating 
the extent to which Islamic thinkers of the period felt 
the need to justify Islam in terms of Western political 
ideals, whether liberal or socialist. 

December 1927 saw the first federation of Indone- 
sian political parties (PPPKI: Permufakatan Perhim- 
punan Poliliek Kebangsaan Indonesia, "Union of Indone- 
sian Political Associations"), subsequent to the 
elimination of the Communist Party after an unsuc- 
cessful localised uprising in 1926. This began a phase 
when there was pressure on all political parties to 
define or re-define themselves in terms of the na- 
tionalist focus indicated by the use of the word "In- 
donesia". Sarekat Islam duly changed its name to Partai 
Sarekat Islam Indonesia ("The Sarekat Islam Party of 
Indonesia"). After Tjokroaminoto's death in 1934, 
PSII lost much of its influence, and from 1937 on- 


wards, a number of breakaway movements left to 
establish separate parties, adding to the plurality of 
Islamic organisations, which already encompassed the 
social movement Muhammadiyah (set up in the same 
year as Sarekat Islam) and the Nahdatul Ulama ("Asso- 
ciation of Ulama") set up in 1926. In an evermore 
repressive colonial situation, it was impossible for any 
movement to replicate the early success of Sarekat Islam 
in mass mobilisation: in 1939, PSII, though still 
larger than any other political party, had only 
12,000 members left. PSII was dissolved at the be- 
ginning of the Japanese Occupation, and in 
November 1943 the Japanese created the Madjelis 
Sjuro Muslimin Indonesia ("Indonesian Muslim Coun- 
cil" or Masjumi) as a single organisational vehicle for 
Islam. This unity lasted until 1952, when the 
Nahdatul Ulama split off, and Indonesia's first elec- 
tions in 1955 were contested by a number of Islamic 
parties which, though they jointly obtained a majority 

f the 

liable tc 

3 form i 

it based on Islam. 
The history of Sarekat Islam is thus paradigmatic of 
that of political Islam in Indonesia, exhibiting a 
tremendous capacity for mass mobilisation, the 
necessity to address powerful competing ideologies 
both traditional and modern, and the effect of contin- 
uing internal divisions in preventing the translation of 
mass appeal into political dominance. 

Bibliography: B. Dahm, tr. P.S. Falla, History of 
Indonesia in the twentieth century, London 1971; Ruth 
Thomas McVey, The rise of Indonesian Communism, 
Modern Indonesia Project, Southeast Asia Pro- 
gram, Cornell University, Ithaca, N.Y. 1965; 
Deliar Noer, The modernist Muslim movement in In- 
donesia 1900-1942, Singapore 1973; H.O.S. 
Tjokroaminoto, Islam dan sosialisme (tertulis di 
Mataram, dalam bulan Nopember 1924); kata 
pengantar [oleh] Hadji Anwar Tjokroaminoto, 
Hadji Harsono Tjokroaminoto [dan] Hadji Ahmad 
Sujud Tjokroaminoto, Lembaga Penggali dan 
Penghimpun Sedjarah Revolusi Indonesia, Djakar- 
ta 1963. (Ann Kumar) 
SARF (a.), a term of Arabic grammar. Literally 
meaning "averting", "divergence", this term 
became in later grammar the current indication for 
the science of "morphology" as a synonym for 
tasrif\q. v.]. In the early stages of Arabic grammar the 
term was used in two completely different senses. In 
Sibawayhi's Kitab, sarf is almost always connected 
with the verb insarafa in the sense of "to be fully 
declined", said of a noun. In Sibawayhi's theory of 
grammar the normal situation for a noun is to have 
three case-endings to indicate the three cases (in 
Western grammars of Arabic such words are said to 
be triptotic in contrast to words that have only two en- 
dings, which are said to be diptotic). This normal 
situation obtains when the noun is as near as possible 
to its lightest, i.e. most flexible form, the triliteral 
masculine noun. Such a noun is said to be completely 
mutamakkin, i.e. having full freedom of movement 
(synonym munsarif). When a noun is diverted from 
this normal pattern it runs the risk of losing its full 
declension. This at least is the way the later gram- 
marians (as for instance al-Zadjdjadj's K. ma yansarif 
wa-ma la yansarif) view the behaviour of these words. 
But it seems that in the Kitab Sibawayhi matters are 
somewhat different; here the emphasis is on the 
absence or presence of nunation. The lightness of a 
word (i.e. its unmarked use with maximum syntactic 
freedom) enables it to carry nunation, but when it 
deviates from its normal pattern, it becomes heavier 
(i.e. more marked) and as a consequence, loses the 
and full declinability (tark alsarf, e.g. Kitab, 

i, 163.11; ii, 2.11 and many other examples, cf. 
Troupeau, 1976, 123, who translates sarf with "con- 
version"). It is difficult to explain this use from the 
lexical meaning of the word, since it is precisely the 
adherence to a certain pattern that brings nunation 
and thus sarf, and only when a word is diverted from 


alladhi I 

awld bihi as al-Khalll s; 
14.5-11, who himself uses the term ma^dul c an in the 
same passage) it loses the sarf (for a detailed analysis 
of the treatment of this issue in the Kitab, see 
Reuschel, 1959, 41-7). 

One explanation for this use of the term sarf may be 
the behaviour of a word like amsi "yesterday". Nor- 
mally, this word has no nunation and only two en- 
dings, since it is used as a temporal adjunct izarf), i.e. 
in a more restricted construction than the syntactically 
free words. But when it is diverted from its usual 
category and used as a proper name it becomes fully 
declinable with nunation (masruf Kitab, ii. 43.6 li-anna 
amsi hahuna laysa c ala 'l-hadd). Perhaps this is the origin 
of the term: sometimes words are diverted from their 
own category to another and then they acquire nuna- 
tion and full declinability. If this explanation is cor- 
rect, the original meaning of the term has been 
generalised to all words receiving nunation (and full 
declinability). In later grammar, the theory of full 
declinability was expressed in a canonical list of fac- 
tors that cause words to lose one of their endings (the 
so-called mawani'- al-sarf): two factors from this list in 
combination (e.g. when a word is both feminine and 
used as proper name) cause a word to become diptotic 
(cf. e.g. in al-Zam akhsh arl's Mufassal, 9-10). 

The discussions in this part of the Kitab are typically 
morphological problems, in which the test of the prop- 
er names is used as a device to find out what the status 
of a word is and to which category it belongs (cf. 
Carter, 1983). These problems have nothing to do 
with the syntactic relations within the sentence, since 
diptotic words, even though they have only two en- 
dings, are syntactically used in all three cases. This 
may explain why in later grammar sarf was used as a 
synonym for tasnf and became one of the normal 
terms for "morphology". 

The lexical meaning of sarafa "to divert, avert" is 
much clearer in the other sense in which sarf is used 
in early grammar, in particular by al-Farra 11 . Here it 
means the divergence or non-identity between two 

al-samaka wa-tashraba 'l-labana, for instance, the sub- 
junctive of tashraba is explained by al-Farra 3 by a prin- 
ciple of "divergence" between the two verbs. This 
construction may be compared to the use of the ac- 
cusative indicating referential non-identity in a 
sentence such as zaydun khalfaka, as against the 
sentence zaydun sdhibuka, where the nominative in- 
dicates the identity of the two nouns. According to 
Carter, 1973, this is Sibawayhi's theory (he uses the 
verb sarafa in connection with this construction, Kitab, 
i, 418-19), but it became associated with Kufan gram- 
mar under the name of sarf, which continued to be us- 
ed in this sense in later explanations of the construc- 
tion (cf. e.g. Ibn al-Anbari, Insaf 229-30; Owens, 
1990, 157 f.; Baalbaki, 1981, 22). The term is, in- 
deed, fairly frequent in al-Farra>'s Ma c ant(e.g. i, 33; 
other instances in Carter, 1973, 298, n. 1). 

Bibliography: Sibawayhi, Kitab, ed. Bulak, 
1316/1898-9; al-Farra 3 , Ma'dm'l-Kur'dn, ed. M. C A. 
al-Nadjdjar, Cairo 1955-72; Ibn al-Anbari, al-Insif 
f! mas&Hl al-khilaf bayn al-nahwiyyin al-basriyyin wa 'l- 
kufiyyin, ed. G. Weil, Leiden 1913; Zamakhshari, 
Mufassal, ed. J. P. Broch, Christiania 1879; al- 
Zadjdjadj, Kitab ma yansarif wa-ma la yansarif, ed. 



H.M. Kara c a, Cairo 1971; R. Baalbaki, Arab gram- 

the third centuries A.D., in Wadad al-Kadi (ed.), 
Studia Arabica et hlamica. Festschrift for Ihsan c Abbas, 
Beirut 1981, 1-26; M.G. Carter, Sarf et hilaf: con- 
tribution a I'histoire de la grammaire arabe, in Arabica, xx 
(1973), 292-304; idem, The use of proper names as a 
testing device in Slbawayhi's Kitab, in C.H.M. 
Versteegh, K. Koernerand H.-J. Niederehe (eds.), 
The history of linguistics in the Near East, Amsterdam 
1983, 109-20; J. Owens, Early Arabic grammatical 
theory: heterogeneity and standardization, Amsterdam 
1990; W. Reuschel, Al-Haltl ibn-Ahmad, der Lehrer 
Sibawaihs, ats Grammatiker, Berlin 1959; G. 
Troupeau, Lexique-index du Kitab de Sibawayhi, Paris 
1976. (C.H.M. Versteegh) 

SARF, a term of Islamic law [see Suppl.]. 
SARHADD (p.), lit. "upper frontier, boundary", 
a general geographical term specifically applied in 
southeastern Persia to the mountain region in the 

Sistan adjoining the frontier with Pakistani 
Balucistan. Its mountain chains run generally from 
northwest to southeast, and include the volcanic (still 
partially active) Kuh-i Taftan (4,042 m/13,262 feet), 
the highest point, but there are also east-west-running 
outliers, such as the Kuh-i Bazman (3,489 m/1 1,478 
feet) which connects the Sarljadd with the Djabal 
Bariz [q.v. in Suppl.]. The only village/small town of 
any note is Kh w ash or Vasht, in a broad and fertile 
valley on the western slopes of the main mountain 
chain and situated on the road linking Zahidan 
(Zahedan) with Iranshahr and Cahbahar on the 
Makran [q.v.] coast. 

The Sarhadd's role in history has been mainly as a 
refuge area e.g. for the Mihrabanid Maliks of Nlmruz 
in the later 9th/15th century (see C.E. Bosworth, The 
history of the Saffarids of Sistan and the Maliks of Nimruz 
{247-861 to 949/1542-3), Costa Mesa, Calif, and New 
York 1994, 463-5). By the later 19th century, the 
region had become largely depopulated, with irriga- 
tion and agriculture abandoned, because of endemic 
banditry; this last provoked in 1916 a punitive expedi- 
tion of the Indian Army mounted from British 
Balucistan into the Kuh-i Morplsh (see R.E.H. Dyer, 
The raiders of the Sarhad, being an account of a campaign of 
arms and bluff against the brigands of the Persia-Baluchi 
border during the Great War, London 1921). 

Bibliography: See also P.M. Sykes, Ten thousand 
miles in Persia or eight years in Iran, London 1902, 93, 
136 ff.; C.E. Bosworth, Historical notes on the Sarhadd 
region in Baluchistan, in Irfan Habib (ed.), Professor 
Mohammad Habib centenary volume, 'Allgafh, forth- 
coming. (C.E. Bosworth) 
SARHANG (p.), a term denoting a rank of of- 
ficer or commander in mediaeval Persian armies 
and paramilitary groups (cf. Vuller, Lexicon persico- 
latmum, ii, 261-2, 293; dux exercitus, praefectus). Thus 
the sarhangs were leaders of bands of c ayyars [q. v. j or 
SunnI orthodox vigilantes combatting the Kharidjls in 
3rd/9th century Sistan, and Ya c kub b. al-Layth, 
founder of the Saffarid dynasty [q.v.], embarked on 
his rise to power by becoming a sarhang in the c ayyar 
forces of a local leader in Bust, Salih b. al-Nadr al- 
Kinani (TaMkh-i Sistan, ed. Bahar, passim; Gardlzi, 
Zayn al-akhbSr, ed. Nazim, 1 1 , ed. Hablbl, 139; see the 
discussion by Bosworth in BSOAS, xxxi [1968], 

In modern Persia, sarahang denotes the rank of 

Bibliography: Given in the article. 

(C.E. Bosworth) 

SARI, Arabic form Sariya, a town of the Cas- 
pian region of Persia, in mediaeval Islamic times 
within the province of Tabaristan, now in the modern 
province of Mazandaran [q.v.) (lat. 36° 33' N., long. 
53° 06' E.). It lies some 32 km/20 miles from the Cas- 
pian Sea on the Tidjin river (Hudad al- c dlam, tr. 77: 
TIzin-Rudh) and in the hot and humid coastal plain; 
the surrounding region has always been famous for its 
silk production and its fruits. 

Whether Sari had any pre-Islamic history is 
unclear, though Islamic lore assigned its foundation to 
the legendary Plshdadid [q.v.] figure, Tahmurath. 
Details are lacking of the first appearance of the Arabs 
there; this may have been in the time of c Uthman's 
governor of Kufa, Sa c Id b. al- c As [q.v.]. Sulayman b. 
c Abd al-Malik's governor Yazld b. al-Muhallab [see 
muhallabids] temporarily occupied the town, but 
was compelled to withdraw by the Dabuyid [see 
dabuya] Ispahbadh Farrukhan. The first Islamic 
building erected there was a congregational mosque 
built by the first c Abbasid governor of Tabaristan, 
Abu '1-Khudayb, in 140/757-8. It remained generally 
the capital of the indigenous Ispahbadh.s, with the 
Arab governors installed at nearby Amul [q.v.], 
although in the 3rd/9th century the Tahirid governors 
and the local 'Alids moved their capital to Sari, and 
the Bawandids remained there till the 7th/ 13th cen- 
tury. The town was flourishing in the 4th/ 10th cen- 
tury, according to the geographers, with a citadel sur- 
rounded by a moat and rampart and a flourishing 
textiles industry. 

There is little mention of Sari in later Islamic times. 
It suffered on Mongol times and was sacked by 
TTmur's troops in 795/1393, yet recovered and re- 
mained the capital of Mazandaran till the Zands, who 
transferred it to Barfurush. Agha Muhammad Khan 
Kadjar brought it back again to Sari, but in the 19th 
century the town declined in favour of Amul and 
Barfurush-Babul and only revived in the 20th century 
with the advent of the railway and of better road com- 
munications during the Pahlavl period. It is now the 
capital both of a shahrastan or canton of the same name 
and of the province (usldn) of Mazandaran; in ca. 1950 
the population was 25,000, but this has now risen to 
185,844 (Preliminary results of the 1991 census, Statistical 
Centre of Iran, Population Division). See further, 

Bibliography: Sam c anl, Ansdb, ed. Haydarabad, 
vii, 128-31; Yakut, Mu'djam al-bulddn, ed. Beirut, 
iii, 170-1; G.N. Curzon, Persia and the Persian ques- 
tion, London 1892, i, 379; Le Strange, Lands, 370, 
374-5; H.L. Rabino, Mazandaran and Astardbdd, 
London 1928, 51-7; Razmara (ed.), Farhang-i 
djughrafiyd-yi Iran-zamtn, iii, 147-9. 

(C.E. Bosworth) 
SARI C , an Arabic metre with the following three 

main types, each consisting of six feet to the line and 

differing in the last foot (darb) only: 






(in which = represents an overlong syllable i. 
sonant + long vowel + consonant). The first and 

fourth foot come in three'variants: u - - ^ u - 

and u - v, -. The second and fifth foot have either 
_ _ „ _ or - ^ o -. In the work of Ibn al-Ruml 
(221-76/836-96 [q.v.]), the distribution of these types 
is; type (a) 6%, (b) 39% and (c) 55%; the three types 
together are used in almost 7% of the poems in the 

scansion of the circle metre is mustafilun mustafilun 
mafildtu (twice). Types (a), (b) and (c) are called 
awwal al-sari', thdni 'l-sarf and thalith al-sari'. Al-Khalll 
b. Ahmad (d. 175/791 [q.v.]) distinguishes three other 
types of sari'; whereas some later metricians mention 
a seventh type. The most important of these are two 
types of three-foot sari': 


r, types (d) and (e) are called radjaz 
[q.v.] by many authorities (cf. Ibn Rashlk, 'Umda, i, 
183; al-Damanhun, Irshad, 76 11. 13-17). The division 
of opinion explains why, for instance, the ten pieces 
in three-foot sari' metres of the Hamdsa of Abu Tam- 
mam are placed under the arajdz in the index of the 
edition of al-Marzuki's commentary, whereas al- 
Tibrlzi's commentary contains an explicit reference to 
the sari' metre in seven of these pieces (see ed. 
Freytag, i, 308, 332, 675, 798, 801, 802, 808; but note 
the fact that the metre of the piece which begins on p. 

297 is 


^ different system is found in al-Djawhari (d. 
393/1003 or later [q.v.]). He does not recognise the 
sari' as an independent metre, but considers the six- 
foot types as sub-forms of a basil from which the 
second foot of each hemistich has been left out ('Arid 
al-waraka, 23-4; Ibn Rashlk, 'Umda, i, 137, ii, 303). 
The three-foot or mashtur types, sc. (d) and (e) above, 
are, according to him, sub-forms of radjaz {'Arid al- 

ii of the grow- 

waraka, 46 n. 15). 

For further statistic: 
ing importance of the metre in c Abbasid 
Bencheikh, Poelique arabe, 203-40. 

Bibliography: G.W. Freytag, Darstellung der 
arabischenVerskunst, Bonn 1830, 242-54; idem (ed.), 
Hamasae carmina cum Tebrisii scholiis integris, Bonn 
1828-51; Muhammad al-Damanhun, al-Irshdd al- 
shdfi 'aid matn al-Kdfi, Bulak 1285/1865; Ahmad b. 
Muhammad al-Marzuki, Sharh Diwdn al-Hamdsa, 
ed. Ahmad Amin and c Abd al-Salam Harun, Cairo 
1371-2/1951-3; Ibn Rashlk, al-'Umda fi mahdsin al- 
shi'r wa-dddbihi wa-nakdih, ed. Muhammad Muhyi 
'1-Dln c Abd al-Hamld, 2 vols., 2 Cairo 1374/1955; 
Abu Nasr Isma'II al-Djawhari, 'Arid al-waraka, ed. 
Muhammad al-*- Alarm, Casablanca 1984; J. E. Ben- 
cheikh, Poelique arabe, 2 Paris 1989. 

(W. Stoetzer) 
al-SARI B. al-Hakam b. Yusuf al-Balkhi, gover- 
nor and financial controller of Egypt from 1 Ramadan 
200/3 April 816. On 1 Rabl< I 201/27 Sept. 816, the 
troops openly mutinied against him, and al-Ma'mun 
was forced to remove al-Sari from his post and replace 
him by Sulayman b. Ghalib; al-Sari was put in prison 
and Sulayman entered upon his office on Tuesday, 4 
Rabl^ I 201/30 Sept. 816. He was removed from of- 
fice as early as 1 Sha c ban 201/22 Feb. 817, as the 
result of a repeated revolt of the troops, and al-Sari 
was again appointed by al-Ma'mun . The news of his ap- 
pointment reached Egypt on 12 Sha c ban 201/4 March 
817; al-Sari was released from prison and entered al- 
Fustat on the same day. He held office till his death 
on 30 Djumada I 205/11 Nov. 820. That al-Sari 
played a prominent part in Egypt even before his ap- 
pointment as governor is evident also from his men- 
tion in the tirdz of a kiswa intended for the Ka c ba of 
the year 197/812-13. His name is also found on gold 
and copper coins of Egypt; see W. Tiesenhausen, 
Monnaies des Khalifes orientaux, 188, no. 1700 (Misr 200 
A.H.), 193, no. 1737 (200 and 202 A.H.); H. Nutzel, 
Katalog d. orient. Miinzen in den Kgl. Museen zu Berlin, i, 
367, no. 2247; Isma c il Ghalib, Meskukdt-i kadime-yi 
isldmiyye katalogl, 188, no. 563 (Misr 200 A.H.), 387, 

no. 928 (Misr 201 A.H.), no. 929 (Misr 204 A.H.). 
Bibliography: KindT, K. al-Wulat, ed. R. Guest, 
London 1912, 161-5, 167-72; Abu *1-Mahasin Ibn 
Taghribirdi, Annates, ed. T.G.J. Juynboll, i, 
Leiden 1855, 574-88; Makrizi, Khitat, i, 178, 179, 
310; Tabari, iii, 1044; Ibn al-Athir, vi, 256; F. 
Wustenfeld, Die Slatlhalter von Aeyplen zur Zeit der 
Chalijen, ii, in Abh. G. W. Gbti. (1875), xx, 30-2; Cor- 
pus papyrorum Raineri III, Series arabica, ed. A. 
Grohmann, i/2, 144^ 145. (A. Grohmann) 

al-SARI b. MANSUR [see abu 'l-saraya]. 
al-SARI b. Ahmad b. al-Sari al-RAFFA 3 al- 
Kindial-Mawsill, Abu '1-Hasan(d. 362/972-3 accord- 
ing to Yakut, Irshad, iv, 185, and Ibn al-<AdIm, 
Bughya, ix, 435; other dates are also given), Arab 
poet and anthologist, particularly famous for his 
descriptive poetry (awsdf). 

He was born in Mawsil, where his father apprentic- 
ed him to the clothes-menders/jobbing tailors 
(rajfd'in), hence his nickname, which is, however, not 
yet used by the contemporary source Ibn al-Nadim 
(Fihrist, 169). In spite of his lowly occupation he tried 
his hand at poetry, and al-Bakharzi lists him among 
poets who had to earn their living by hard work, 
alongside the rice-bread baker al-KhubzaruzzT, the 
market-crier al-Wa'wa 3 al-Dimashki, and the cloth 
merchant Abu Hilal al- c AskarI [q.uo.] (Dumyat al-kasr, 
ed. al-Hulw, Cairo 1968, i, 528-9). He was apparent- 
ly not a complete autodidact, since the name of his 
teacher, the otherwise unknown Abu Mansur Ibn Abi 
Barrak, is mentioned by Ibn al-Nadim (Fihrist, 169, 1. 
12), but he is otherwise said to have been uneducated. 
After achieving a certain fame as a poet, he went to 
the Hamdanid court in Aleppo in 345/956-7 (for the 
year, see Ibn al- c Adim, Bughya, ix, 428) and found ac- 
ceptance among the court poets of Sayf al-Dawla. 
After the latter's death in 333/945, he moved on to 
Baghdad and sang the praises of the Buyid vizier al- 
Muhallabi (d. 352/963 [q.v.]). The sources are not in 
unison about his worldly affairs in this period: some 
depict him as poor and debt-ridden (al-Khatib al- 
Baghdadi, Ta\ikh Baghdad, ix, 194), others as wealthy 
(al-Tha c alibI, Yatima, ii, 119; Yakut, Irshad, xi, 185). 
For the time of al-Muhallabl's successor al- c Abbas b. 
al-Husayn al-Shirazi (appointed vizier in 356/967, d. 
362/973 [q.v.]) we have an anecdote related by al- 
Muhassin b. Ibrahim al-Sabi 5 (d. 401/1010; see al- 
sabi, no. 8), according to which al-Sari must have 
been a man of modest means at the time (apud Ibn al- 
c Adim, Bughya, ix, 429-30). 

The main feature of al-San's life as depicted in the 
sources is his constant feud with the two Khalidl 
brothers, Abu Bakr Muhammad b. Hashim and Abu 
c Uthman Sa c id b. Hashim (see al-khalidiyyan). 
They too came from Mawsil and it seems that already 
early on al-Sari accused them of stealing his own as 
well as other poets' poetry. Ibn al-Nadim, in his 
paragraph on the Khalidis. concurs that it was second 
nature to them to appropriate (ghasabd) verses they lik- 
ed (Fihrist, 169, 1. 21-3). To make the accusation stick, 
al-Sari used the devious method of making copies of 
the diwdn of Kushadjim (d. ca. 350/961 [q.v.]), whom 
he admired and emulated, and including the best 
poems of the Khalidis in them (al-Tha c alibi, Yatima, 
ii, 118; Yakut, Irshad, xi, 183; Ibn Khallikan, ii, 360). 
The feud between al-Sari and his adversaries was a 
long-lasting one; in the account of al-Khatib al- 
Baghdadi, the Khalidis appear as al-Sari's ineluctable 
nemesis, who succeed in poisoning his relationships 
with Sayf al-Dawla, al-WazIr al-Muhallabl, and other 
patrons. Al-Sari himself was, however, not immune to 
the accusation of sarika: Ibn al-Nadim— amidst a 


string of admiring epithets — calls him a man "of 
many thefts" {kathir al-sarika, Fihrist, 169, 1. 28). It is, 
however, important here to distinguish between two 
kinds of plagiarism, (a) wholesale lifting of other 
people's poems {musdlaia, for the term see e.g. Yatima, 
ii, 119,1. 5), and (b) taking up, and playing with, ex- 
isting and attributable motifs. Both play a role in 
al-Sari's literary life. On the one hand, Ibn al-Nadim 
mentions that al-Sari is said to have appropriated the 
poetry of his teacher Ibn Abi Barrak (Fihrisl, 169, 1. 
12), and al-Tha c alibI writes that he discovered a 
number of identical poems in the collection of poetry 
of the two Khalidls in the handwriting of Abu 
'Uthman, the younger of the two brothers, and the 
collection of al-Sari's poetry in the poet's own hand- 
writing, both belonging to the bibliophile Abu Nasr 
Sahl b. al-Marzuban {Yatima, ii, 118-19; one of the 
poems he quotes in this regard is actually also includ- 
ed in Kushadjim, Diwdn, ed. Khayriyya M. Mahfuz, 
Baghdad 1390/1970, 230, no. 210, and has even a 
fourth avatar in an unattributed appearance in al- 
Sari's own anthology K. al-Muhibb [see below], iv, 
326, no. 717). Al-Tha c alibl declares himself unable to 
decide whether the duplication of poems is due to 
tawdrud "two poets having the exact same idea" or 
musdlaia "appropriating another poet's poems" 
{Yatima, ii, 119, 11. 4-5). If theft is involved, it is still 
unclear who is the thief and who the victim. In his 
chapter on the Khalidi brothers, al-Tha c alibI thus 
uses the term tasdruk "two-way theft" to characterise 
the cases of strong similarity between the Khalidls and 
al-Sari which he presents to the reader {Yatima, ii, 
184-6 [Abu Bakr], and 199-201 [Abu c Uthman]). He 
also makes an important observation concerning the 
reason that made the free floating, or malicious steal- 
ing, of so many poems possible: there is, between al- 
Sari and the Khalidls. "an amazing agreement and a 
close similarity in handling the reins of the rhymes 
and fashioning the adornment of the motifs" {Yatima, 
ii, 119, 11. 5-6). The only way of redress that a poet 
had to combat an infringement of his "copyright" 
was to complain {lazallum) about this injustice to the 
authorities, most commonly to the recipients of their 
panegyrics. A fair amount of al-Sari's poems contain 
such complaints against the Khalidls. 

The other type of plagiarism is the more regular 
and — in an age of mannerist poetising — hardly 
reprehensible type of adopting and, if possible, im- 
proving successful motifs of earlier poets. Al-Tha c alibl 
gives a list of 44 such cases and praises al-Sari for his 
skilful "plagiarism" (husn al-sarika wa-djawdat al- 
a khdh . see Yatima, ii, 120, penult.). The "victims" are 
almost all "modern" poets, led by al-Mutanabbl who 
serves as a model twelve times. 

Al-Sari al-Raffa 3 has left three works, two of which, 
the Diwdn and al-Muhibb wa 'l-mahbub wa 'l-mashmum 
wa 'l-mashrub, have been preserved. The latter is an 
anthology of the ma'am'-catalogue type, i.e. a topically 
arranged selection of poetic fragments with descrip- 
tions of the various parts of the beloved, of the lover, 
of spring, and of wine. The majority of the poets are 
"Moderns", but Umayyad poets have their fair 
share, while pre-Islamic ones occur very rarely. 

The Diwdn was, according to Ibn al-Nadtm {Fihrist, 
169, penult.), edited by the poet himself shortly before 
his death on ca. 300 leaves who then added more ma- 
terial to it. It was also edited by an unknown 
"modern" adib in the alphabetic arrangement accord- 
ing to rhyme letters. This may mean that the poet's 
edition was not alphabetical. If so, the situation may 
be reflected in the surviving 
which are alphabetical, while others 

description of the mss. in Diwdn, i, 188-208). How- 
ever, the situation was very likely more complicated: 
al-Tha c alibf enumerates three sources for his 
knowledge of al-Sari's poetry: the Diwdn as brought to 
him from Baghdad, a number of outstanding poems 
recited to him and given for copying by the famous 
poet and epistolographer Abu Bakr al-Kh w arazmi (d. 
383/993 [q.v. under al-kh w ArazmI]), and a volume 
(mudxallada) in al-Sari's own hand-writing in the 
possession of Abu Nasr Sahl b. al-Marzuban and con- 
taining many additions to the Diwdn. But still, he 
says, he found in the secondary literature fragments of 
poems by al-Sari that he could not trace in his sources. 
Ibn al- c Ad!m lists nine people, among them Abu Bakr 
al-Kh w arazmi, who are said to have transmitted 
al-Sari's poetry {Baghya, ix, 428). The details of the 
early transmission of al-Sari's poetry are thus still 

The opinions about al-Sari's stature as a poet were 
generally very high. He is even called "next in line" 
{radif) after al-Mutanabbl and superior to him in 
tenderness {rikka) (on the title-page of the Diwdn ms. 
Laleli 1745, see introd. of ed. in Diwdn, i, 189). He 
is particularly famous for his ecphrastic poetry, "very 
versatile in similes and descriptions" {kathir al-iftindn 
ft 'l-tashbihdt wa 'l-awsdf), as Ibn al-Nadim says 
{Fihrist, 169, 29). The index of poetically described 
objects in the recent Diwdn edition lists 150 items (!). 
Not all of them yield monothematic poems, many are 
integrated into larger frameworks. Some of these 
seem to be unusual polythematic structures, possibly 
created by al-Sari. Thus poem no. 353 {Diwdn, ii, 
464-5) starts with a description of the morning and a 
rooster, leads on to a wine song, and ends with a 
panegyric on Sayf al-Dawla. An interesting subgenre 
with al-Sari is his descriptions of his own poetry, of 
which al-Tha c alibi gives a small collection of examples 
{Yatima, ii). A close literary study is still lacking. 

Bibliography: Works. Diwdn, first ed. Cairo 
1355 (based on two mss.); ed. Habib Husayn al- 
Hasanl, 2 vols., [Baghdad] 1981 (based on ten 
mss.); al-Muhibb wa 'l-mahbub wa 'l-mashmum wa 'l- 
mashrdb, ed. Misbah GhalawundjI [?], 4 vols. [vol. 
4, ed. Madjid Hasan al-Dhahabl], Damascus 1306- 
7/1986 (the ms. Vienna 359 bearing the same title 
is not al-Sari's work; the ed. has identified it as a 
rearranged version of Halbat al-kumayi by al- 
Nawadji [q.v.], see introd., m4-m5). 

Studies. Yusuf Amln Kasir, al-Sari al-RajfiP, 
Baghdad 1956; Habib Husayn al-Hasanl, as an in- 
trod. to his ed. of the Diwdn, i, 9-182; for com- 
parisons with the ecphrastic poetry of Kushadjim, 
see Alma Giese, Wasfbei Kusdgim, Berlin 1981 , 222 
(snow), 255 (melon), 263 (candle), 269 (fire), 273 
(writing utensils). On K. al-Muhibb see J. Sadan, 
Maidens' Hair and Starry Skies. Imagery systems and 
ma' i ani guides; the practical side of Arabic poetics as demon- 
strated in two manuscripts, in Israel Oriental Studies, xi 
(1991), 57-88, in particular 67-70, 74-84. 

Sources. Ibn al-Nadim, al-Fihrist, ed. Flugel; 
Tha c alibi, Yatimat al-dahr, ed. M.M. c Abd al- 
Hamid, 4 vols., 2 Cairo 1375-7/1956-8; al-Khatlb 
al-Baghdadi, Ta^rikh Baghdad, ix, Cairo 1349/1931; 
Yakut, Irshddal-arib, ed. A.F. Rifa'I, xi, Cairo n.d.; 
Ibn Khallikan. Wafaydt al-a'-ydn, ed. I. c Abbas, ii; 
Ibn al- c Adfm, Bughyat al-talab fi ta\ikh Halab, facs. 
ed. F. Sezgin, ix, Frankfurt 1409/1989; Muhsin al- 
Amln, A c ydn al-ShFa, xxxiv, Beirut 1370/1950, 
35-147. _ (W.P. Heinrichs) 

SARi al-SAKATI, Abu '1-Hasan b. al-Mughallis, 
important Sufi of the second generation of Sufis in 
Baghdad, 155-253/772-867. 



Born as the son of a pedlar (sakati) who had settled 
at an early date in the Karkh [q.v.] quarter of 
Baghdad, Sari rose to be a distinguished wholesale 
trader, known for his honesty (Ta~>rikh Baghdad ( = TB) 
ix, 189). Like other merchants, he devoted himself to 
hadith studies which, as the names of his teachers in- 
dicate, must have brought him from Baghdad via 
Kufa to Mecca (Hilya, x, 127). 

At the age of ca. 35-4-0 he encountered the saint 
Ma c ruf al-Karkhr (d. 815 [q.v.]), which brought his 
career as a merchant to a sudden end and initiated the 
second period of his life. The world had beome indif- 
ferent to him; he abandoned his trade and chose the 
Sufi path. Several journeys opened up new horizons 
for him. In the convent of 'Abbadan [q. v. ] he tried to 
join the Basra school through fasting exercises. On the 
road he came to know the Syrian anchorite C A1T al- 
DjurdjanT, who exerted a lasting influence upon him 
{Hilya, x, 110-11) and directed his eye towards Syria, 
where he became acquainted with the school of 
Ibrahim b. Adham [q.v.], which was to impress him 
in more than one way. He stayed at Damascus, 
Ramla, Jerusalem and Tarsus in the north, from 
where he joined the djihad, at already sixty years of age 
(Hilya, x, 126). His wandering years ended around 
218/833 with his definite return to Baghdad. Probably 
already before his journeys he had won there Bishr al- 
Haff (d. 84-0 [q.v.]) as a paternal friend. For his views 
he was also indebted to Fudayl b. c Iyad (d. 803 [q. v. ]), 
whom he did not, however, come to know personally. 
He does not seem to have maintained close relations 
with al-Muhasibi (d. 857 [q. v. ]), who was very closely 
connected to him as far as Sufi ethics and self-educa- 
tion were concerned (cf. J. van Ess, Gedankenwelt, 10). 

The third period of Sari's life, which began im- 
mediately after his return to Baghdad, was that of his 
great role as a teacher. Pupils did not only come flock- 
ing in from c Irak and Khurasan (Abu '1-Kasim al- 
Djunayd, Abu Sa c fd al-Kharraz, Abu '1-Husayn al- 
Nuri, Samnun, Ibn Masruki and others), but also 
from Syria (e.g. c Ali al-Ghada'iri and Isma c fl b. c Abd 
Allah al-Shami). It was not only professional Sufis 
who joined him, but also pious workmen and mer- 
chants, unto the muhaddith [see hadith] and the kadi 
(TB, ix, 191-2, 189, 1. 15). Sari must have been an 
outstanding teacher, who knew how to address and 
stimulate everyone, whether personally or in the 
course of lectures (e.g. Hilya, x, 119, 11. 15 ff.), but 
who also knew what he could demand of the in- 
dividual person. Only the intimates had insight in his 
mystical experience. On the other hand, he also com- 
posed a rule for novices (Hilya, x, 117 below). 

We do not know how long Sari was active as an 
educator. From a certain moment onwards he 
withdrew completely into his own world of experience 
and permitted only intimates, above all Abu '1-Kasim 
al-Djunayd (d. 910), to approach him. For this fourth 
period of Sari's life, too, people in Baghdad had a 
concrete explanation. Once he was teaching in his cir- 
cle when the misanthrope Abu Dja'far al-Sammak 
passed and marred the holding of further sessions by 
remarking: "Abu '1-Hasan, you have become the 
resting-place of the idlers" (TB, xiv, 411). 

him, Sari appears as a noble, unselfish, helpful and 
dynamic personality who, as a member of the Islamic 
community, felt himself responsible for the well-being 
of his fellow men (cf. TB, ix, 188). This attitude had 
already won him the favour of Ma c ruf al-Karkhl and 
made him later responsive to the juluwwa [q.v.] of 

Ibrahim b. Adham, who did every kind of work for 
others but would himself never lay claim to their help. 
The after-effects of his influence also sharpened Sari's 
sensitivity to decent good manners (adab, cf. Hilya, x, 
120, 11. 20-1, 122, 11. 18-20). 

Like Ibn Adham, Sari was a man of action and not 
a theoretician. Knowledge (Him) interested him only 
in so far as it could be turned into deed (<ama[). He 
shared Bishr al-Hafl's scepticism of knowledge of the 
hadith; according to him, it was "no provision for the 
hereafter" (ibid., x, 127). Consequently, he transmit- 
ted only traditions which were illustrated by his own 
attitude or which would support the latter. He was no 
less at a loss with theology, even if occasionally he 
knew how to gain a Suff aspect from a theologumenon 
(cf. Mukhtasar Ta\ikh Dimashk = MTD, ix, 227 in the 
middle). There was only one knowledge for which he 
had a burning interest, namely, the paths to self- 
knowledge and self-education; as a protagonist of this 
field of knowledge, he particularly esteemed al- 
Muhasibf (Kit al-kulub, ii, 35, 11. 11-12 = Gramlich, 
Die Nahrung, i, 503). 

Other main features of Sari's character were his 
sincerity and veracity. Bishr al-Haff had trained him 
in the freedom not to fool others in anything (Luma'- 
373, 11. 5-8), and Sari himself demanded that nothing 
should be done or left out "on behalf of people" but 
only on behalf of God alone (Sifat al-safwa, ii, 213), in 
other words, that there should be practised what in 
c Irak was understood as ikhlds ]q.v.] and which Sari, 
in association with al-Djurdjam, still called lashih al- 
irada "purification of intention" (al-Sulami, 51; for 
al-Djurdjani, see Hilya, x, 112). In the method of car- 
rying through this attitude against all impulses of 
hypocrisy (riya*), he largely followed al-Muhasibi (cf. 
ibid., x, 125, and van Ess, Gedankenwelt, 14-9 ff.). 

Above all, he fought a particular case of hypocrisy, 
'hich one might qualify as "business by means of 

. On a lowt 


thanks to one's religion" (Hilya, x, 117, II. 7-10; s 
for this, Luma 1 , 201, 1. 3; Meier, Abu SaHd, 304); on 
a higher one, he censures the claim to social authority 
of many a Kur'an reader on the ground of his piety 
and religious knowledge, which he misuses (MTD, ix, 
226-7). As a general remedy and as preventive 
medicine, he recommends "anonymity" (khumuC), 
following also in this point Bishr al-Haff' s example. 

Sari belongs to the generation of Sufis in which — 
not least under the influence of a new mystical 
experience — the authentic wealth of ideas of zuhd is 
partly commuted, partly refined, but sometimes also 
hyperbolically brought ad absurdum. The theatrical 
tirades of the ascetics against the world became silent, 
and the fierce battle against its allurement gave way 
to indifference (see for this, MTD, ix, 219). Sari in- 
deed abandoned his trade because of lack of interest 
in wordly affairs, but on the other hand he did not 
simply throw away all his worldly goods but put them 
into action for works pleasing to God. Zuhd became an 
inner attitude and signified that "the heart is free 
from that from which the hands are free" (Luma 1 , 46). 

The most important means of commuting zuhd was 
"scrupulousness" (wara°), i.e. the painful effort to 
come only in contact with "what is permitted" (haldl). 
The first matter to be dealt with was one's daily 
bread. The scrupulous person wanted to know where 
his sustenance came from, and who had dealt with it 
before it reached his hands. Sari then extended this 
method to all implements of daily life (TB, ix, 189). 
He apparently came to know these implements in 
Syria, in the heritage of Ibn Adham (cf. Hilya, x, 1 16, 


22-3). The idea of narrowing what is permitted by 
means of wara c in such a way that zuhd practically loses 
its object goes also back to this heritage. This was in 
any case the way in which Sari proceeded (TB, ix, 
190). When sure of their irreproachable origin, he was 
by all means open to worldly titbits (MTD, ix, 217 
below). But he followed Ibn Adham's ideas of wara c 
also in another aspect; not only did he not want to eat 
anything which might burden him before God, but 
neither should food render him indebted to any 
creature (Luma c 183, 11. 6-7; on Ibn Adham, cf. al- 
Kushayri, 59, 11. 9-10 = Gramlich, Das Sendschreiben, 

All along there had been a latent conflict between 
excessive service of God (Hbdda [q. v. ]), as practised by 
the ascetics, and social requirements. Initially, Sari 
had clearly decided in favour of the latter. He 
transmitted to his pupils (al-Sulaml, 165, 11. 9-10) the 
relevant hadith, taken over from Ma c ruf: "He who 
fulfills a wish of his Muslim brother is rewarded in the 
same way as one who has served God during all his 
life" (Manakib, 626-7). In the last period of his life, 
however, he recommends as "the direct path into 
paradise" that "one should occupy oneself with the 
service of God, turn only to Him in such a way that 
there does not remain anything else inside oneself 
(Hilya, x, 119). Ascetic piety expressed in works is cer- 
tainly not at stake any more here, but it was on the 
one hand a matter of the delight of being alone with 
God (cf. ibid., x, 117, 1. 14, 125, 1. 10) and on the 
other a matter of avoiding temptations, no less impor- 
tant for Sari (al-Sulaml, 50, 11. 6-7). 

With the conviction of his own depravity, Sari 
enters the field of hyperbolism. This conviction may 
have been furthered by i 
by examination of his 
responds to an old praxis of exercise in humility (cf. 
van Ess, Gedankenwelt, 158): one forces oneself to the 
notion that one is worse than everybody else. Fudayl 
b. c Iyad [q.v.] held the same view (Hilya, viii, 101). 
To the alarm of his pupils, Sari did not even wish to 
have preference over homosexual boys (al-Sulaml, 49, 
11. 10 ff.) and continuously squinted at the tip of his 
nose in order to investigate whether his face had not 
already become black because of his depravity (Hilya, 
x, 116; this passage became famous, cf. Ta'-arruf, 31, 

San also gives a new turn to another ascetic motive, 
namely ostentatious and continuous mourning as it 
was practised by the bakka^un and later by Fudayl b. 
c Iyaj. Sari still allows for this attitude when he men- 
tions "weeping about one's sins" as the first of the 
"five most beautiful things" (al-Sulami, 54). Later, 
being mournful becomes for him the characteristic of 
someone who loves God (as in Hilya, x, 125, 1. 24), 
and when he exclaims before al-Djunayd "I should 
wish that the mourning of the entire mankind be 
thrown upon me" (ibid., 118); this may conceivably 
have an altruistic meaning, but certainly not the 

Finally, a word on the primal ascetic motive of fear 
of God and His Judgement. It is true that 
reminiscences of both the absolute and the polarised 
fear of God are not lacking in Sari's thinking (e.g. 
Hilya, x, 118, 11. 2-4, and al-Sulaml, 53, 1. 5), but the 
starting point of his reflections is a well-considered 
balance between fear and its counter-force, namely, 
hopefulness (for this problem see Meier, Abu SaHd, 
148 ff.), put down in a fragment of a letter of a later 
period, the only authentic document of Sari which we 
possess (Lurna', 238). However, he developed this 
dichotomy further in a Sufi way. On the one hand, he 

does this with the help of a surmounting third notion: 
firstly, it is the shame (haya 7 ) before God's eye (Hilya, 
x, 117), which takes the place of the mere fear of His 
punishment and hope for His grace (see also c Attar, 
Tadhkira, i, 253, 11. 14-15, 247, 11. 20 ff.); secondly, 
Sari introduces here as a fourth binding force, after 
the example of Shaklk al-Balkhl (Adab al Hbadat, 20-1), 
the love of God, which outshines the elementary ex- 
periences of fear and hope. On the other hand, he 
refines fear and hope into feelings of reverence (hayba) 
and intimacy (wis) (Sifat al-safwa, ii, 215, 11. 7-8). All 
"new" notions are characterised as personal feelings 
towards God and therefore indicate that we find 
ourselves already in the sphere of mysticism. In the 
"five things next to which nothing else rests in the 
heart", Sari gives a summing up, namely "fear of 
God alone, hope in God alone, love of God alone, 
shame before God alone, and intimacy with God 
alone" (Hilya, x, 124-5). 
Sari's mystical experiences 

We are extremely badly informed about what Sari 
made public of his mystical experience (cf. al-Sulaml, 
48, 11. 3-2; al-Kushayrl, 10,1.2 = Gramlich, Das Send- 
schreiben, 41). It is certain that this experience was 
completely under the imprint of the love of God. The 
impulse might go back to Ma c ruf al-Karkhl who, after 
his death, could only be imagined in paradise as being 
drunk with the love of God (Manakib, 675-6) and who 
is said to have been moved by the love of God to that 
form of renunciation of the world which he also knew 
how to provoke in Sari (Kit, iii, 82, 11. 4 ff.). In fact, 
it is to al-Djunayd exclusively that we owe all informa- 
tion on Sari's love of God, and so it dates only from 
the third or fourth periods of his life. 

Here, too, in the most authentic field of his 
mysticism, Sari is seen as the practical man who is 
averse from all forms of speculation. He does not want 
to have anything to do with the Sufi theory of love; on 
the contrary, he wants to understand love of God as 
a psychosomatic phenomenon. While speaking about 
this, he tried to extend the skin of his forearm, and 
when this proved to be impossible, he said, "If I were 
to assert that this skin had dried up on this bone out 
of love for Him, I would speak the truth" (al- 
Kushayri, ii, 1. 11 = Gramlich, Das Sendschreiben, 41- 
2). Elsewhere, too, he depicts the love of God as an in- 
ner burning (TB, ix, 191, 11. 7-10) and does not get 
tired of quoting corresponding verses on profane love 
(Liana', 251, 11. 4-6). It is, therefore, no wonder that 
he also transmitted the famous-infamous hadith al-Hshk 
(al-Kushayrl, 1 12, 1. 23 = Gramlich, Das Sendschreiben, 
317; for the hadith itself, see Massignon, Essai, 195-6). 

Sari represents here the well-known conviction that 
lovers of God would be tested by Him in the hardest 
possible way, so that the truthfulness of their love 
might prove itself. In a dream he hears God telling 
them: "I want to bring down upon you as many 
proofs as you have breaths, which not even the solid 
mountains would stand. Will you bear them patiently 
(a-tasbiruna)?" They answer: "If it is You who tests 
us, do whatever You wish!" (Sifat al-safwa, ii, 216, 11. 
15 ff.). Sari himself did not advance any further opin- 
ion on this, but we possess a vivid description by al- 
Djunayd of the torments in question, where he depicts 
the desperate endeavours with which the mystic tries 
to win back the situation of bliss, caused by the union, 
after he has awaken from ecstasy (Kitab al-Fand'^i- 
8 = Gramlich, Islamische Mystik, 19-22). Perhaps the 
linguistic virtuoso that the pupil was, retained here his 
master's experiences. 

Sari was indeed very well acquainted with the 
ecstatic heights of mystical experience. Al-Djunayd 


>ut a trance of the master, in which his face 
[ such a way that the bystanders could no 
tand the sight (Luma c , 307, 11. 13 ff.). More- 

kiredji (prh 

r, Sari confirmed tc 

n that 


•^idhddda) and intensive thoughts of God (adhkdr 
kawiyya) could lead the mystic so far that "his face 
might be beaten with a sword without him being 
aware of it" {ibid., 306, 11. 6-12). It may be that what 
Sari once indicated as his nightly "experiences" 
(fawd'id) {Hilya, x, 121, 11. 20-1) were such states of 
trance, and that what he called his awrdd [see wird] 

sidered ecstasy as a charisma {kardma), he did not of 
it (cf. his postulate of the 


!, HUya, 

120, 1 

Bibliography: 1. Sources . (a) The most reliable 
are TaMkh Baghdad; AbG Nu c aym al-Isfahanl, 
Hilyat al-awliyd\ Cairo 1351-7/1932-8; SulamI, 
Tabakat al-sufiyya, ed. Shurayba, Cairo 1372/1953; 
Kushayrl, Risdla, Cairo 1359/1940, tr. R. 
Gramlich, Das Sendschreiben al-Qusayris uber das 
Sufitum, Wiesbaden 1989; Ibn al-DjawzI, Si/at al- 
safwa, Haydarabad 1388-92/1968-72. (b) Sources 
with weaker traditions are Abu Talib al-Makkl, Kut 

kulub, Cai: 


lich, Die 

ten, Stuttgart 1992-4; Sarradj, K. al 
Luma c fi 'l-lasawwuf, ed. Nicholson, London-Leiden 
1914; KalabadhI, al-Ta<arruf li-madhhab ahl al- 
tasawwuf, ed. Arberry, Cairo 1934; Ibn Manzur, 
Mukhtasar TaMkh Dimashk, ed. Nashshawl, 
Damascus 1405/1985. (c) Unreliable sources are 
Hudjwirl, Kasjtf al-mahcgub, ed. Zhukovski, Lenin- 
grad 1926; c Abd Allah Ansarl, Tabakat al-sufiyya, 
ed. Hablbl, Kabul 1340 ASH; c Attar, Tadhkirat al- 
awliyd\ Tehran 1336 ASH. (d) Other sources are 
Shaklk al-Balkhl, Adah al-Hbdddt, ed. P. Nwiya, in 
Nusus sufiyya ghayr manshura, Beirut 1986; Djunayd, 
Kitdb al-Fand\ ed. A.H. Abdel-Kader, in The life, 
personality and writings of Al-Junayd, London 1976; 
Manakib MaHij al-Karkhi wa-akhbdruh, ed. in al- 
Mawnd, ix/4 (Baghdad 1401/1981), 609-80. 

2. Studies. G. Bowering, The mystical vision oj ex- 
istence in Classical Islam, Berlin 1980; J. van Ess, Die 
Gedankenwelt des Hdnt al-Muhdsibi, Bonn 1961; R. 
Gramlich, Islamische Mystik, Sufische Texte aus zehn 
Jahrhunderten, Stuttgart 1992; L. Massignon, Essai 
sur les origines du lexique technique de la mystique 
musulmane, 2 Paris 1954; F. Meier, Abu SaHd-i Abu 'l- 
Hayr, Leiden 1976; P. Nwyia, Exegese coranique et 
langage mystique, Beirut 1970; B. Reinert, Sari al- 
Saqati und seine Bedeutung fur die islamische Mystik, in 
Onens, xxxv, forthcoming. _ (B. Reinert) 

man of letters and bureaucrat (?992-?1071/? 

He was also reported to have been a good 
calligrapher and an ardent lover and cultivator of 
flowers, Ibrahim I dubbing him seTshukufedji (cicekci 
bash!) (see Omer Faruk Akiin, in IA, art. San Ab- 
dullah). In his own works he is referred to as c Abd 
Allah b. al-Seyyid (or al-Sherlf) Muhammad b. c Abd 
Allah (Akiin, loc. at.). He seems to have been born in 
Istanbul, but the sources disagree on the date of birth. 
His father Sayyid Muhammad had fled from the 
Maghrib to Istanbul and settled there; his mother was 
a daughter of the Beglerbegi Mehmed Pasha (d. 
997/1588-9), brother of the Grand Vizier Khalll 
Pasha, by whom he was brought up and given an 
education under Shaykh Mahmud of Scutari and 
others (see Nihad Sami Banarh, Resimli Turk edebiyati 
tanhi, Istanbul 1976, 700), with early exposure to 
mysticism, Khalll Pasha himself having close tarika 
connections [see khalil pasha kaysariyyeli]. 

te secretary) to Khalll 

given command of the troops in the Persian cam- 
paign, and in 1037/1627-8 was appointed re^is ul-kuttab 
[q.v.] in place of Mehmed Efendi, who had just died, 

After the latter's death in 1038/1629, he remained out 
of office until 1040/1638 when he was appointed to the 
imperial rikdb [q.v.]. He accompanied Murad IV on 
his Baghdad campaign and then became re^is ul-kiittdb 
for the second time. He continued in various positions 
until 1065/1655 when he retired from public life and 
devoted his remaining years to writing. His tomb is in 
the cemetery of Maltepe outside the Top Kapi (Gate 
of St. Romanus) in Istanbul (Gibb, HOP, iv, 79). 

A member of the Bayramiyya tarika, Sari c Abd 
Allah Efendi had a close relationship also with leaders 
of the Mawlawiyya [q.v.], and most of his works (in 
Turkish and Arabic) were connected with mysticism. 
His five-volume Djewdhir-i bewdhir-i Methnewi, written 
1035-41/1625-31 and dedicated to Murad IV, is a 
Turkish translation of and commentary on the first 
volume of Mawlana's Persian Mathnawi, but has been 
described as an "encyclopaedia of mysticism" (Akiin, 
loc. cit.), containing as it does information about vari- 
ous orders and legends of saints. The interpretation is 
closely related to Sari c Abd Allah's knowledge of Ibn 
al- c Arabi's [q.v.] doctrines. Although stylistically the 
work is part of the high culture, with sections in rhym- 
ed prose, it also contains sentences that are short and 
simple, and close to the spoken language (Banarh, op. 
cit.). It was printed at Istanbul in 1288/1871. Among 
his other works are: Nasihatu 'l-muluk, a Mirror for 
Princes and discussion of death and the hereafter; 
Themerdtu 'l-ju^ad, a discussion on mystical topics with 
Isma c Il Ankarawl [q.v.], which also refers to various 
orders and personages; Djawharalu 'l-biddya, another 
mystical work but dedicated to Murad IV, with an in- 
troduction describing his recent conquest of Baghdad; 
and Dusturu 'l-insha', a collection of official cor- 
respondence and other documents (some of his own 
composition), .interesting from the aspect of foreign 
relations. He also wrote poetry and songs of a mystical 
tenor under the makhlas c AbdI. 

Bibliography. For further details and bibl., see 

Akiin, op. cit. 

. .. (Cl. Huart-[Kathleen Burr.ll]) 

SARI KURZ or Sar! Gorez (on the pronunciation 
of the second element in the name, see M.C. 
§ehabeddin Tekindag, Yeni kaynak ve vesikalar isigi 
altinda Yavuz Sultan Selim'in Iran sejeri, in Tarih Dergisi, 
xvii/22 [1967], 49-78, n. 20), lakab [q.v.] of the Ot- 
toman scholar and statesman Nur al-DIn Hamza 
b. Yusufof Karasi. 

The date of his birth is unknown. According to 
Tashkopru-zade [q.v.] (al-ShakaHk al-nu'mdniyya fi 
<ulamd> al-dawlat al-'Uthmdniyya, Beirut 1395/1975, 
181), he studied with the "<ulama> of his age", and 
successively with Khatlb-zade and Khodja-zade 
[q.v.]. He next "entered the service" of Khodja Sinan 
Pasha [q.v.], Grand Vizier between 881/1476 and 
882/1477 (I.H. Uzuncarsih, Hizir Bey oglu Sinan 
Pasa 'nin vezir-i dzamligina ddir cok kiymetli bir vesika, in 
Belleten, xxvii [1963], 37-44), accompanying his 
patron to exile in Sivrihisar in 882/1477 (Tashkdprii- 
zade, op. at., 106-9). On his accession in 886/1481, 
Bayezid II [q.v.] recalled Sinan Pasha, and appointed 
him muderns at the Dar al-Hadlth in Edirne [q.v.], 
whither Sari Kiirz accompanied him as his teaching 
assistant. After the Dar al-Hadlth, Sari Kiirz became 
muderris "at several madrasas" , finally at one of the 
Eight Madrasas attached to the Mosque of Mehemm- 
ed II [see sahn-i thaman]. He evidently remained on 


intimate terms with Bayezid II, since the only incident 
in his career to enter the Ottoman historiographical 
tradition is his embassy on behalf of this sultan to his 
son Prince Selim (Sellm I [q.v.]) in 917/1511 (Die 
Altosmanischen Anonymen Chroniken, ed. F. Giese, 
Breslau 1922, 130; Ahmet Ugur, The reign of Sultan 
Selim I in the light of the Selim-ndme literature, Berlin 1985, 
160, 171, 177; Djelal-zade Mustafa, Selim-ndme, ed. 
Ahmet Ugur and Mustafa Quhadar, Ankara 1990, 
258, 295). Bytheyear917/1511, Bayezid had also ap- 
pointed him kadi of Istanbul, a post which he held un- 
til 919/1513 or later (for wakfiyyes which he validated 
in these years as kadi of Istanbul, see O.L. Barkan and 
E.H. Ayverdi, Istanbul oaktflari tahrir defeteri. 935 
(1546) tarihli, nos. 2107, 906). Selim I appointed him 
successively kadi '■asker [q.v.] of Anatolia and kadi '■asker 
of Rumelia. It was he who in 920/1514 issued the fal- 
wd sanctioning the opening of hostilities against the 
Safawids [q.v.] and the massacre of their adherents in 
Anatolia (for text and facsimile of the fatwd, see 
Tekindag, op. cil.). Tashkopru-zade reports that 
Selim eventually dismissed him as kadi '■asker of 
Rumelia "because of something that happened be- 
tween them", and re-appointed him to one of the 
Eight Madrasas. He seems to have ended his career as 
kadi of Istanbul, since he validated wakfiyyes in Istan- 
bul in 924/1518 and 927/1521 (O.L. Barkan and E.H. 
Ayverdi, op. cit., nos. 917, 2163). He died, 
Tashkopru-zade says, "in 928 or 929"/1521-3, and 
was buried "next to his mosque in the city of Con- 
stantinople". The mosque in question must be the 
mosque of "Sari Kez", which Ewliya Celebi [q.v.] 
(Seydhat-ndme, i, Istanbul 1314/1896-7, 310) places "in 
the C A1I Pasha market, near [the mosque of Mehemm- 
ed] the Conqueror". The mosque gave its name to a 
quarter in the Fatih district of Istanbul, known 
variously as Sari Kez or Sari Giizel (Semavi Eyice, 
istanbul'da yayla camileri, in Tarih Dergisi, x [1954], 34 

Tashkopru-zade records Sari Kurz as the author of 
a work of Hanafl/uru' entitled al-Murtadd (see HadjdjI 
Khalifa, ed. Fliigel, Kashf al-zunun, v, 491), and a 
book of responsa to the legal conundrums of Sayyidi 
Hamidl (ibid. , vi, 241). HadjdjI Khalifa (iv, 170) also 
lists a gloss on al-Isfahanl's commentary on the 
Tawdli c al-anwdr of al-BaydawI. 

Bibliography: In addition to the references given 
in the text, see R.C. Repp, The Mufti of Istanbul, 
Oxford 1986, 218-20; Mehmed Thiireyya, Sidjill-i 
<Othmdni, repr. Farnborough 1971, iv, 581; Haflz 
Hiiseyn Aywansarayl, Hadikat al-djawdmi'-, Istanbul 
1281/1864-5, 133-4. (C. IMber) 

SARI MEHMED PASHA, Bakkal-zade, Hadj- 
djI, ShehrI, Defterdar (?-l 129/1717), Ottoman 
statesman, born in Istanbul (hence: Shehri), a son of 
a Muslim Turkish grocer. He styled himself El-Hddj^ 
Mehmed ed-Defteri in the preface to his chronicle. 

He made his career in the financial department of 
the Porte [see maliyye]. In 1081/1671 he was 
employed in the office of the riizndmdje-yi ewwel kalemi 
[q.v.]. He won the patronage of the defterdar Kilic C A1I 
Efendi, bashdefterddr(m function 1102/1691-2) and was 
promoted to mektubdju ( = mektubi), head of the 
secretariat of the principal defterdar. The Grand Vizier 
Rami Mehmed Pasha [q. v. ] appointed him principal 
defterddr (17 Dhu '1-Hidjdja 1114/5 May 1703). This 
was the first of his seven appointments to this high of- 
fice. At the time of the Edirne Revolt he was dismissed 
(13 Radjab 1115/23 August 1703). The new Sultan 
Ahmed III [q. v. ] reinstated him soon at the behest of 
the rebels, who demanded Sari Mehmed's financial 
acumen to produce the necessary funds for their "Ac- 

lon Fee" (djulus bakhshishi). Dismissed after this 
ation, he was relegated to the ruzndmdje-yi ewwel 
e. Defterdar again on 23 Shawwal 1115/29 
February 1704, he lost the position after eleven 
months. Re-appointed 14 months later, he remained 
in function for one year and 5 months. The next turn 
of office lasted from 20 Dhu '1-Ka<da 1119/12 
ebruary 1708 till 3 Djumada II 1121/10 August 
709, when he lost favour with the Grand Vizier 
iorlulu '■All Pasha [q. v. ] and retired into private life 
t his konak in the Kumkapi quarter of the capital. The 
Grand Vizier removed him from the centre of power 
by appointing him (titular) mutasarrif of Salonica, 
ranking as a pasha of "two tails" [see tugh] and 
sending him as military governor to the frontier for- 
i town of Bender (Bessarabia) [q. v. ] . Later, he was 
n the government of Izmid as well. Sari Mehmed 
Pasha became defterddr again in 1123-4/1712 for five 
nths. After the conclusion of peace with Russia at 
rne (1125/1713), he was appointed a member of 
boundary commission which settled the new 
border in Podolia. Upon their return to Istanbul, the 
sultan awarded caftans of honour to inter alios Sari 
Mehmed, who soon was made defterddr again 
(1126/1714). He appears to have been a client of the 
kapuddnpasha Ha djdj I Mehmed Pasha "Djanim 
Khodja" [q.v.]. Next year he joined the campaign in 
the Morea led by the Grand Vizier Damad 'All Pasha 
[q.v.]. His task was the provisioning of the army via 
Egriboz [q.v.]. After the Grand Vizier's death in the 
battle of Peterwardein, Sari Mehmed Pasha received 
the rank of vizier (4 Ramadan 1128/22 August 1716). 
He had been hoping for the grand vizierate, but he 
loyally served under the new Grand Vizier, the aged 
HadjdjI ArnawOd Khalil Pasha [q.v.], fighting Prince 
Eugene of Savoy and the Imperial army near 
Belgrade. He seems not to have become member of 
the faction of the sultan's favourite, Damad 
Newshehirli Ibrahim Pasha [q.v.]. Sari Mehmed 
openly vented his disappointed ambition and lost the 
confidence of the sultan. In 1129/1717 he was recalled 
> the Porte, and his new appointment as military 
governor of Salonica meant his disgrace. He received 
the order to organise a force of 3,000 to be recruited 
from the Ewldd-i Fdtihdn (originally Anatolian Turks 
settled in Rumelia) at his own expense and to join the 
main army at Nish to take up the guarding of the 
"Iron Gate" of the Danube. He failed to comply, and 
openly criticised the sultan's policy. His behaviour led 
to his downfall; he was accused of incitement to revolt 
and of previous misconduct in the field before 
Temesvar in 1716, where his force of 1000 dalkllic- 
serdengecti volunteers withdrew without fighting. A 
kapidjtbashi sent from the Porte had him executed in 
the castle of Kawala [q.v.]. His possessions in Istan- 
bul, Salonica, Bender and Siwas were confiscated. At 
least one son survived him, Mehmed Emln, alaybeyi at 
Jiwas in 1179/1767 (see above, vol. I, at 295a). 

Bibliography: Nasd^ih ul-wuzerd'' we'l-iimerd* weyd 
Kitdb ul-Guldeste, ed. and tr. W.L. Wright, Ottoman 
statecraft. The book of counsel for vezirs and governors, 
Princeton 1935; ed. in mod. Tkish, with an introd. 
by H.R. Ugural, Defterdar San Mehmet Pasa, devlet 
adamlanna bgutler, Ankara 1987; a chronicle of Ot- 
toman history 1656-1704, pop. ed. in mod. Tkish. 
A. Ozcan, Zubde-i vekayiat (olaylann bzu), Istanbul 
1977-9, 3 vols. (vol. 4 still to come?), a critical ed. 
by idem, announced in 1990; idem, Defterdar San 
Mehmed Pasa'mn mali bazi gorus ve faaliyetleri, in 
GDAAD, x-xi (1981-2 [publ. 1983]), 239-48; i. Par- 
maksizoglu (ed.), Silahdar Mehmet Aga. Nusretndme, 
Istanbul 1962-9, 3 vols., in 5, ii, index, s.v. Mehmed 


Efendi, Bakkaloglu; A. Tabakoglu, Gerileme ddnemine 
girerken osmanit maliyesi, Istanbul 1985, 132, 230, 
266; i.H. Uzuncarsih, Osmanh tarihi, iv/2, Ankara 
1959 v 597-602, (A.H. de Groot) 

SARI SALTUK DEDE, Turkish warrior-saint 
of the 7th/13th century. 

The sources on his life are extremely meagre. The 
earliest surviving work in which he is mentioned is 
Ibn Battuta's Rihla. The author states that when he 
passed through the town of "Baba Saltuk" in the 
Dasht-i Kipcak [q.v. in Suppl.] (probably near the 
lower Dnieper in the Ukraine) in 732/1332, he learn- 
ed that its namesake was "an ecstatic devotee, 
although things are told of him which are reproved by 
Divine Law" (tr. H.A.R. Gibb, The travels of Ibn Bat- 
ma, Cambridge 1958-71, ii, 499-500). About the 
same time, legendary stories related to Sari Saltuk had 
entered the Bektashl wildyet-names. By the 9th/15th 
century, he was considered to be a Bektashl saint 
(F.M. Koprulu, The Seljuks of Anatolia, tr. and ed. G. 
Leiser, Salt Lake City 1992, 81 n. 82). Next, he ap- 
pears briefly in the TaMkh-i dl-i Saldjuk, written by the 
Ottoman historian Yazidji-oghlu (or Yazidji-zade) 
C A1I during the reign of Murad II (824-55/1421-51). 
According to this work, which is virtually the only 
historical source on his life, Sari Saltuk "of blessed 
memory" went to Constantinople to join the deposed 
Saldjuk sultan c Izz al-Dln Kaykawus II and his army 
after they had taken refuge in Byzantine territory 
from Kaykawus's brother and his Mongol protectors. 
These Turks found favour with the Byzantine 
emperor, who subsequently gave them Dobrudja 
[q.v. ] as an abode. Later, the emperor imprisoned 
Kaykawus, but he was freed by Berke Khan of the 
Golden Horde, who gave him and his followers 
hospitality in the Crimea. Berke Khan then transfer- 
red the Turks of Dobrudja, among them Sari Saltuk, 
into the steppe (dashl). When Kaykawus died in the 
Crimea in 679/1280, one of his sons claimed the 
throne and asked for permission to return to Rum. At 
that point, Berke Khan ordered Sari Saltuk to lead the 
remaining Turks back to Dobrudja. Meanwhile, an- 
other of Kaykawus's sons had been held in Constan- 
tinople, where he was baptised by the Patriarch. Sari 
Saltuk successfully asked for him to be freed. He 
returned to Islam and became a dervish. Sari Saltuk 
then transferred to him the supernatural power that 
he had received as a shepherd from the holy man 
Mahmud KhayranI (d. 667/1269) of Akshehir and 
sent him to Sultaniyya. Sari Saltuk later died in 
Dobrudja (P. Wittek, Yazidjtoghlu '-Alton the Christian 
Turks of the Dobruja, in BSOAS, xiv [1952], 639-68). 
While describing Siileyman's Mohacs campaign 
(932/1526), Kemal Pasha-zade [q.v.] (or Ibn Kamal, 
d. 940/1534), Ta>rtkh-i dl-i c Olhmdn, mentions Sari 
Saltuk in passing as a wonder-working saint in 
Dobrudja. Sayyid Lukman (d. ca. 1010/1601-2 [q.v.]), 
Idjmdl-i ahwal-i al-i Salajuk, essentially repeats Yazidji- 
oghlu's information, but adds that Sari Saltuk went to 
Rumelia in 662/1263-4 (relevant passages from these 
writers in A. Decei, Le problem de la colonisation des Tuns 
seljoukides dans la Dobrugea au XIII' si'ecle, in Tarih Aras- 
tirmalan Dergisi, vi [1968], 85-111). 

Yazidji-oghlu's report clearly reflects the incident 
in which Kaykawus II [q.v.] fled from the Mongols to 
the court of Michael VIII Palaeologus in 660/1262 
and subsequently reached the Crimea. Some of the 
Turks who followed him later settled permanently in 
Dobrudja and were thus apparently the first, to 
establish Islam in the Balkans. Nevertheless, most of 
these Turks eventually became Christians known as 
the Gagauz [q.v.]. We have neither a contemporary 

istian (Byzantine) nor Muslim source, however, 
mentions a Sari Saltuk in connection with these 
Us. It is also noteworthy that none of the aforesaid 
Ottoman historians links Sari Saltuk with the 
Bektashls. Yet by the 9th/15th century, Sari Saltuk 
appears in Bektashl tradition as a shepherd whom 
Hadjdji Bektash sends, via Sinope, to Georgia, where 
he converts the Georgians. Then he goes to Dobrudja, 
where he kills a dragon which had captured the 
daughters of a king, at the fortress of Kaliakra 
(Kilghra). Afterwards, he calls the people to Islam 
and builds a tekke (Koprulu, The Seljuks, 54-5). 

Relying primarily on Bektashl wildyet-names and 
other legendary or semi-legendary sources, it is 
generally held today that there actually was a saint 
named Sari Saltuk and that he was a disciple of 
Hadjdji Bektash. Moreover, the heterodox dervish 
Barak Baba [q.v.] claims, in turn, that he was a disci- 
ple of Sari Saltuk (Koprulu, Islam in Anatolia, tr. and 
ed. G. Leiser, Salt Lake City 1993, 22-3). Indeed, 
Koprulu considers Sari Saltuk to be one of a series of 
warrior babas or alp-erens connecting the uprising of 
Baba Ishak [q. v. ] in eastern Anatolia in 638/1 240 with 
the revolt of Badr al-DIn [q. v. ] b. Kadi Samawna in 
Dobrudja in 819/1416(o/). ctt, 15,22, 77 n. 55, 78 n. 
57, 90 n. Ill, and esp. A.Y. Ocak, La revolle de Baba 
Resul ou la formation de I'heterodoxie musulmane en Anatolie 
au XIII' Steele, Ankara 1989, 100-5 and passim). Still, 
it is unclear if this Sari Saltuk and the one who accom- 
panied the Turks to Dobrudja in the 7th/13th century 

■e the s; 

e pers. 

, by the mid 9th/15th century, it was the 
legendary Sari Saltuk who, as an heroic figure in the 
epic Saltuk-ndme, supplanted the vague historical per- 
sonage in the Turkish consciousness. In 878/1473 
when Prince Djem [q.v.] was in Edirne guarding the 
Balkan borders while Mehemmed II was on campaign 
against Uzun Hasan, he heard many stories about 
Sari Saltuk from various places in Rumelia and com- 
missioned one Abu '1-Khayr al-Ruml to compile a 
book about him. Abu '1-Khayr collected material for 
seven years, visiting all the places associated with Sari 
Saltuk in Anatolia and Rumelia, and then wrote his 
Saltuk-ndme. This work followed, but was superior to, 
the Battdl-ndme and Ddnishmend-name as the last in the 
cycle of epic romances concerning the conquest of 
Anatolia and Rumelia. The first two works were cen- 
tred on Anatolia, while the Saltuk-ndme focussed on 
Rumelia. In this epic, Sari Saltuk was a great Sufi 
who had close relations not only with Hadjdji 
Bektash, but also with Djalal al-Dln al-Ruml, Abu 
Ishak al-Kazarunl and even Nasr al-Dln Khodja 
[q.vv. ]. He commanded miraculous powers and 
defended Muslims and converted unbelievers from 
China to Andalusia. He first lived in Sinope, then in 
the Crimea, then along the Danube, and finally in 
Edirne. (He was often associated, initially, with 
northern Anatolia. The Bolu sdl-nime of 1334/1915-6, 
226, for example, states that the region of Bartin 
north of Bolu near the Black Sea had previously been 
referred to as "Saltuk ili", the province of Saltuk.) 
His major objective was to Islamise Rumelia, and he 
predicted the conquest of that region in the time of the 
sons of c Othman. Although the Saltuk-ndme is replete 
with fabulous elements (including stories common to 
those in the Bektashl wildyet-names), and blends 
numerous local traditions (he was often identified with 
such Christian personalities as St. Nicholas) with 
Muslim or Turkish traditions, it also reflects many 
historical events that occurred between 1200 and 
1400, such as the struggle of the Anatolian Saldjuks 
and beyliks against the Mongols, Byzantines and other 


Western powers; the relations of the Golden Horde 
with the West and Byzantium; the struggle of the 
Aydinid ruler Umur Bey with the Christians of the 
West; and the establishment of the Ottoman State. 

The Saltuk-name was written, in fact, much like the 
first anonymous Ottoman chronicles. It is especially 
important for describing the psychological state of the 
unbelievers created by the Ottoman invasion of the 
Balkans (Koprulii, The Seljuks, 43-52). It even sheds 
light on social life in Anatolia under the Ilkhans and 
in the Crimea under the Golden Horde. 

Ewliya Celebi (d. ca. 1095/1684 [q.v.\) reports the 
existence of two works, now lost, on the legendary 
Sari Saltuk: a risdla by Yazidji-oghlu Mehmed Celebi 
(d. 855/1451) and a Saltuk-name by Ken'an Pasha (d. 
1069/1659 [q.v.]) (Seydhat-ndme, Istanbul 1896-1935, 
iii, 366/Topkapi Sarayi Bagdat Koskii ms. 305, fol. 
127b). Ewliya claims that his real name was Muham- 
mad Bukhari, that he was girded with a wooden sword 
by Ahmad Yasawi [q. v. ] and sent to the assistance of 
Hadjdji Bektash, who then sent him to Dobrudja; that 
he lived in Arpa Cukuru, Siwas and Tokat; and that 
he was the patron saint, pir, of the boza makers (op. 
cil., ii, 134/Bagdat ms. 304, fol. 266a, and i, 
659/Bagdat ms/304, fol. 212b). Ewliya also relates 
the story that Sari Saltuk instructed his disciples to 
bury him in seven coffins in remote towns in Rumelia 
(indeed, as far away as Sweden) "so that the ig- 
norance of where the body really was would produce 
everywhere a pilgrimage of Muslims and from the 
pilgrimage would result the incorporation of these 
lands into the kingdom of Islam" (op. cit., ii, 
133/Bagdat ms. 304, fol. 266a ff.). This indicates, of 
course, that the legendary Sari Saltuk was not only a 
major symbol of the Islamisation and Turkification of 
Rumelia, but that he was also probably a composite 
of many of the warrior dervishes who went to that 
region after the original Sari Saltuk's death. Certain- 
ly, the fact that most of the followers of the original 
Sari Saltuk eventually converted to Christianity is 
strong evidence that the legendary person was much 
more important and influential in this respect than the 
original. In addition to his burial sites in Rumelia, he 
had resting places as well in Anatolia. The latter are 
revered even today: Iznik, Bor near Nigde, and 
Diyarbakir (G. Smith, Some turbeslmaqdms of San 
Salluq, an early Anatolian Turkish gdzi-saint, in Turcica, 
xiv [1982], 216-25). Nevertheless, his "true" burial 
place is generally considered to be at Babadaghi [q.v.], 
in Dobrudja just south of the Danube delta. This site 
was often the centre of Turkmen, ghazi and heretical 
dervish activity. While on campaign, Bayezld II 
visited Babadaghi in 889/1484 and built there a great 
mosque and zawiye dedicated to Sari Saltuk. Accord- 
ing to Ewliya, he also rebuilt the saint's turbe, which 
had become an object of pilgrimage, and endowed a 
madrasa and baths in the town. Later, in 945/1538, 
Siileyman also showed interest in Sari Saltuk and 
spent several days visiting his tomb. In the 12th/18th 
century, Babadaghi began a long period of decline. 
But the saint's modest tomb still stands, an often- 
repaired and humble reminder of a thriving Ottoman 
past (J . Deny, San Saltiq et le nom de la ville de Babadaghi, 
in Melanges offerts a Emile Picot, Paris 1913, 1-15; M. 
Kiel, The turbe of Sari Saltik at Babadag- Dobrudja, in 
Guney-Dogu Avrupa Arastirmalart Dergisi, vi-vii [1977-8], 

Bibliography (in addition to works cited in the 
text): M.F. Koprulii, Turk edebiyyatlnda ilk 
mutasawwlflar, Istanbul 1918, 2nd revised ed. 
Ankara 1966, 54-5; F.W. Hasluck, Christianity and 
Islam under the sultans, Oxford 1929, 429-39, index; 

A.Z.V. Togan, Umumi tiirk tanhine giris, Istanbul 
1946, repr. 1981, 268-71, 334-5, 342-3; O. 
KSpriilu, Tarihi kaynak olarak XIV ve XV. asirlardaki 
bdzi Tiirk menakibndmeleri, diss. Istanbul University 
1951, Tarih Semineri Kutiiphanesi, no. 512; O. 
Turan, Selcuklular zamamnda Tiirkiye tarihi, Istanbul 
1971, 499, 581; F. iz, facsimile ed. Saltuk-name, 
Cambridge, Mass. 1974-84; S.H. Akalin (ed.), 
Saltuk-name, Ankara 1987-90; K. Yiice, Saltuk- 
name 'de tarihi, dini ve efsanevi unsurlar, Istanbul 1987, 
with full bibl. (G. Leiser) 

SARIKA (a.), theft, noun of agent Sarik "thief. 
Islamic legal theory distinguishes between al-sarika 
al-sughrd (theft) and al-sarika al-kubra (highway robbery 
or brigandage), each with different hadd punishments. 
(1) Theft (sarika) is punished by cutting off the 
hand, according to sura V, 42. This was an innova- 
tion of the Prophet's; but, according to the Awa^il 
literature, this had already been introduced in the 
days of paganism by al-Walid b. Mughira (Noldeke- 
Schwally, Gesch. d. Qprdns, i, 230). This method of 
punishment may be of Persian origin (cf. Lettrede Tan- 
sar, ed. J. Darmesteter \nJA, Series 9, iii [1894], 220- 
1, 525-6; Sad Dar, 64,5 = Sacred books of the East, xxiv, 
327). In pre-Islamic Arabia, theft from a fellow- 
tribesman or from a guest was alone considered 
despicable, but no punishment was prescribed for it; 
the person had himself to see how he could regain his 
property (G. Jacob, Altarabisches Beduinenleben 2 , 217- 
18; cf. J.L. Burckhardt, Bemerkungen uber die Beduinen, 
Weimar 1831, 127 ff. 261 ff.). In the beginning of the 
lst/7th century, the right or left hand was cut off; 
there was no fixed rule. The Kur'an leaves the point 
obscure, and one tradition says that Abu Bakr ordered 
the left hand to be cut off (Muwatta > Sarika, bdb 4; al- 
Shafi c i, Kitdb al-Umm, vi, 117). Cf. the variant of sura 
V, 42, aymdnahumd, transmitted by Ibn Mas c ud. 

According to the teaching of the fukahd^, the thief's 
right hand is cut off (for a second crime the left foot, 
then the left hand, then the right foot) and at the wrist; 
the stump is held in hot oil or fire to stop the bleeding. 
The Hanafis and Zaydls, however, put the culprit into 
prison at his third crime, which the Shafi c is and 
Malikis only do after his fifth. The ShT c Is inflict im- 
prisonment for the third offence and death for the 
fourth. The punishment was inflicted in public; the 
thief was frequently led round the town seated 
backwards on an ass with the limb cut off hung round 
his neck (cf. Ibn Madja, Hudad, bdb 22; O. Rescher, 
Studien uber den Inhalt von 1001 Nacht, in 1st. [1919], ix, 
68 ff.). Punishment could not be inflicted in cases of 
pregnancy, severe illness or when the weather was 
very cold or very hot. It is a hadd punishment, as a 
right of God (hakk Allah) is violated by theft. But as the 
rights of the owner are also injured (hakk adami) the 
thief is bound to make reparation. If the article stolen 
has disappeared, he is kept under arrest (not so, ac- 
cording to Abu Hanifa). The caliph c Umar is said 
always to have condemned the thief to return double 
the value (cf. Roman Law: Justinian, Instit., 4, 1, 5). 
The jurists define theft for which the hadd punish- 
ment is prescribed as the clandestine removal of legal- 
ly recognised property (mat) in the safe keeping (hirz) 
of another of a definite minimum value (nisdb; among 
the Hanafis and Zaydis 10 dirhams, among the 
Malikis, Shafi c is and Sh^is 'A dinar or 3 dirhams) to 
which the thief has no right of ownership; it is so dis- 
tinguished from usurpation (ghasb) and embezzlement 
(khiydna). By hirz is meant guarding by a watchman or 
by the nature of the place (e.g. a private house). Thus 
theft from a building accessible to the public (e.g. 
shops of the market, in the open air, baths) is not 


liable to the hadd punishment. This is further only ap- 
plied to one who (1) has attained his majority (baligh 
[q. v.}), (2) is compos mentis ('akil) and (3) has the inten- 
tion (niyya) of stealing, i.e. is not acting under compul- 
sion but freely (mukhtar). No distinction is made be- 
tween freeman or slave, male or female. The 
punishment is not applied in case of thefts between 
husband and wife and near relatives nor in the case of 
a slave robbing his master or a guest his host. Views 
are divided on the question of the punishment of the 
dhimmi and the protected alien (musta^min) with the 
hadd, and on the punishment of accomplices and ac- 

ter must reach the nisab for each of the thieves. It is 
not theft to take articles of trifling value (wood, water, 
wild game) and things which quickly go to waste 
(fresh fruit, meat and milk), or articles in which the 

which are not legitimate articles of commerce (mil), 
like freeborn children, wine, pigs, dogs, chess-sets, 
musical instruments, golden crosses (the theft of a full- 
grown slave is considered ghasb) or articles in which 
the thief already has a share (booty, state treasure, 
wakf, something from the common good to the value 
of the share), also copies of the Kur'an and books (ex- 
cept account books), as it is assumed the thief only 
desires to obtain the contents. The conception of 
literary theft is unknown to fikh. 

The charge can be made by the owner and 
legitimate possessor (or depository) but not by a 
second thief. The legal inquiry has to be conducted in 
the presence of the person robbed. For proof two male 
witnesses are necessary or a confession (ikrar [q.v.], 
which can, however, be withdrawn. It is recommend- 
ed to plead not guilty if at all possible [see c adhab[. If 
the thief, however, has given back the article stolen 
before the charge is made, he is immune from punish- 
ment (sura V, 43). 

(2) Highway robbery or robbery with violence 
(muharaba, kaf al-tarik) occurs when anyone who can 
be dangerous to travellers falls upon them and robs 
them when they are distant from any possible help or 
when someone enters a house, armed, with the inten- 
tion of robbing (cf. Roman Law; Justinian, Novellae, 
134, ch. 13). The Shl<is consider any armed attack, 
even in inhabited places, as highway robbery. The 
same regulations hold regarding the person and the 
object as above, especially the nisab. On the authority 
of sura V, 37-8, the culprit is liable to the following 
hadd punishments. If a man has committed a robbery 
which is practically a theft to be punished with hadd, 
his right hand and left foot are cut off (the next time, 
the left hand and the right foot). If, however, he has 
robbed and killed, he is put to death in keeping with 
right of reprisal (kisds) and his body publicly exposed 
for three days on a gibbet or in some other way. The 
punishment of death is here considered a hakk Allah; 
the payment of blood-money idiya) is therefore out of 
the question. If the criminal repents, however, before 
he is taken, the hadd punishment is omitted; but the 
claim of the person robbed of the article for compensa- 
tion and the talio remain. All accomplices are punish- 
ed in the same way; if one of them cannot be held 
responsible for his actions, the hadd punishment can- 
not be inflicted on any. 

All these laws hold only for the hadd punishment 
which the judge can only inflict when all conditions 
are fulfilled. In all other cases the thief is punished 
with ta c zir [q.v.] and condemned to restore the article 
or to make reparation. It is the same with the thief 
who comes secretly but goes away openly (mukhtalis) 
or the robber who falls upon someone and robs him 

at a place where help is available {muntahib). Special 
laws were therefore frequently passed in Islamic states 
to supplement the SharFa, in Turkey, for example, by 
the Ottoman sultans Mehemmed II (Mitteilungen zur 
Osm. Gesch., i [1921], 21, 35), Suleyman II (von Ham- 
mer, Staatsverfassung, i, 147-8). Mehemmed IV and 
c Abd al-MedjId. These laws endeavoured more and 
more to replace the hadd punishment by fines and cor- 
poral punishment. The Turkish criminal code of 1858 
still only recognised fines and imprisonment for theft, 
although the SharPa was not officially abolished 
thereby [see medjelle). 

The punitive prescriptions of the SharFa regarding 
sarika, which have been either abolished or largely 
mitigated during the course of the 20th century, ex- 
cept in such countries as Saudi Arabia and the Sudan, 
seem likely to regain ground at the dawning of the 
third millennium in several parts of the Islamic world 
with the rise of Islamic fundamentalism. 

Bibliography: The chs. on sarika and kaf al-tarik 
in the fikh books, to which should be added: H. 
Laoust, Le precis de droit d'Ibn Qudama ( = Fr. tr. of 
the K. al-Wmdafiahkam al-fikh by Muwaffak al-DTn 
Ibn Kudama, d. 620/1223), Beirut 1950, 266-9. See 
further, Th.W. Juynboll, Handbuch des islamischen 
Gesetzes, 305-6; E. Sachau, Muhammedanisches Recht, 
825 ff.; L.W. van den Berg, Beginselen van het 
Moham. Recht\ Batavia-The Hague 1883, 189-90 
(cf. Snouck Hurgronje, Verspr. Geschrijten, Bonn 
1923, ii, 196-7); Keyser, Het moham. Strafiegt, The 
Hague 1857, 11-12, 101-2, 161-2; D. Santillana 
(tr.), Sommario del dintto di Haiti, Milan 1919, ii, 
724 ff.; A. Querry, Droit musulman, Paris 1872, ii, 
514 ff.; Tornauw, Moslem. Recht, Leipzig 1855, 
236; W. Heffening, Islamische Fremdenrecht, Hanover 
1925, 15, 28-9; E. Nord (tr.), Das tiirkische 
Strafgesetzbuch von 1858 mil Novelle von 1911, Berlin 
1912, arts. 62 ff. and 216 ff.; Young, Corps de droit 
ottoman, 1906, vii; van den Berg, Strafiecht der Tiirkei, 
in F. von Liszt (ed.), Die Strafgesetzgebung der Gegen- 
wart, 1894, 710 ff.; Jaenecke, Grundprobleme des 
tiirkischen Strafrechts, Berlin 1918; L. Bercher, Les 
delits et les peines de droit commun prevus par le Coran, 
Tunis 1926; A.F. Bahnasi, al-DJard^im fi 'l-fikh al- 
islami, Cairo 1959; J. Schacht, An introduction to 
Islamic law, Oxford 1964, 175-6, 179-81; N.J. 
Coulson, A history of Islamic law, Edinburgh 1964, 
124, 150, 157. See also fikh; hadd; and esp. 
mahkama and the Bibl. there. (W. Heffening) 
SARIYA [see sArI]. 
SARKAR AKA (p.), a term used for a number 

Shi*! tradition. It appears to have originated in the 
19th century, possibly in recognition of links between 
the title's bearers and the Kadjar court. The title 
(meaning something like "lord and chief) was used 
for the first Aka Khan (Hasan C A1I Shah, 1804-81 
[q.v.] and several of his successors, as_ heads of the 
NizarT Ismaffis (sometimes as Sarkar Aka Khan); it 
is, however, not in current use. Leaders of the 
Shaykhi branch of the Twelver gh/a [see 
shaykhiyya] have been termed "Sarkar Aka" since 
the time of Hadjdj Mirza Muhammad Karfm Kir- 
mam (1809-70 [q.v.], as referred .to as "Sarkar-i 
Khan"), a great-nephew of Fath- C A1I Shah and a son- 
in-law of Nasir al-Dfn Shah. The title passed to his 
Kirman-based lineal successors within the IbrahTmT 
family until quite recently, and was particularly used 
of the late Abu '1-Kasim Khan (d. 1979). Within the 
Baha 3 ! movement [see baha'is] the title is reserved for 
'Abbas Effendi c Abd al-Baha' (1844-1921), the son of 
Mirza Husayn 'All Nun Baha 3 Allah [q.v.], whose 



family had a variety of marital links to the Kadjars. In 
English usage, Baha'Is refer to him as "the Master", 
which is both a partial translation of Sarkar Aka and 
an echo of Christian terminology. (D. MacEoin) 

SARKHAD [see salkhad]. 

SARKHEDj, a site 10 km/6 miles to the southeast 
of Ahmadabad [q.v.] in western India, capital of the 
sultans of Gudjarat [q.v.] in the 9th- 10th/ 15th- 16th 

Its fame arises from the complex of buildings built 
round an artificial lake, all of them still standing and 
excellent specimens of 9th/15th century Gudjarat ar- 
chitecture. They include the tomb of the saint Shaykh 
Ahmad Khattu "Gandj Bakhsh" (d. 850/1446) and a 
mosque of sultan Muhammad Shah (846-55/1442- 
51). It became a favourite retreat of Sultan Fath Khan 
Mahmud Shah Begf a (862-917/1 458-1511), who built 
the large tank, palace buildings and two mausolea for 
himself and his family. 

Bibliography. J. Burgess, The Muhammadan ar- 
chitecture of Ahmadabad A. D. 1412-1520 (= Archaeol. 
Survey of Western India, vii), i, 46-51 and pis. 9, 55- 
64; M.S. Commissariat, A history of Gujarat, 
London-Bombay 1938-57, i, 131-3, 235-7; K.A. 
Nizami and M. Habib (eds.), A comprehensive history 
of India. V. The Delhi Sultanate (A. D. 1206-1526), 
Delhi, etc. 1970, 876; J. Burton Page, Mosques and 
tombs, in G. Michell and Snehal Shah (eds.), 
Ahmadabad, Bombay 1988, 30-119, and pis. on 
48-55. (Ebba Koch) 

SARLIYYA, the name of a group of KakS'is or 
Ahl-i Hakk [q.v.] living in northern c Irak, in a group 
of six villages, four on the right bank of the Great Zab 
and two on its left one, not far from its confluence 
with the Tigris and 45 km/28 miles to the south-south- 
east of Mawsil. The principal village, where the chief 
lives, is called Wardak, and lies on the right bank; the 
largest village on the left bank is Sufayya. 

The Sarlls, like the other sects found in northern 
c Irak (Yazidis, Shabaks, Badjuran), are very uncom- 
municative with regard to their belief and religious 
practices, so that the other inhabitants of the country 
have in the past attributed abominable rites to them 
and alleged that they have a kind of secret language 
of their own. In 1902, Pere Anastase gave some notes 
on the Sarlls (and also on the sects of Badjuran and the 
Shabaks) which he obtained from an individual in 
Mawsil. According to him, their language was a mix- 
ture of Kurdish, Persian and Turkish. As to religion, 
they were monotheists, believing in certain prophets, 
in paradise and hell. They neither fasted nor prayed. 
They believed that their chief had the power to sell ter- 
ritory in paradise. For this purpose he visited all the 
villages at harvest time, and every Sarll was allowed 
to purchase as many dhird's as he could pay for; the 
price of a dhird'- was never less than a quarter of a 
medjidiyye. Credit was not granted. The chief gave a 
receipt which show how many dhird^s an individual 
had acquired. This receipt was put in the pocket of the 
dead man so that he could present it to Ridwan, the 
guardian of Paradise. The Sarlls had also a feast-day 
once in every lunar year, which consisted in the con- 
sumption of a repast at which the chief presided, and 
to which every one contributed a cock boiled with rice 
or wheat. After this meal, called aklat al-mahabba, the 
lights are said to have been extinguished and orgies of 
promiscuity to have taken place. The head of the com- 
munity was succeeded at his death by his unmarried 
son; he was forbidden to shave his beard or his 
moustache. The Sarlls were polygamous. They were 
said to have a sacred book written in Persian. 

These statements should be taken with considerable 

reserve. The Sarlls themselves said that they were 
simply Kurds and belonged originally to the Kake 
Kurds who have some villages near Kirkuk. But the 
Kake Kurds also had a mysterious reputation. A 
characteristic feature noticed in one of the Sarll 
villages (Sufayya) was an ornament with triangular 
holes in the walls of the principal buildings of the 
village. Like the Yazidis [q. v.], the Sarlls used Muslim 
names, and their chief in the early part of this century 
was one Taha Koca or Mulla Taha. 

The present (1994) status, or even the continued 
existence, of these Sarlls, is unknown, given the pres- 
ent impossibility of western scholars undertaking 
ethnological field work in the Kurdish areas of 
northern c Irak. 

Bibliography: V. Cuinet, La Turquie d'Asie, 
Paris 1894; Pere Anastase al-Karmill, Tafkihat al- 
adhan ft la c rif lhalathal adyan, in Machriq, v (1902), 
577 ff.; W.R. Hay, Two years in Kurdistan, London 
1921, 93-4; C.J. Edmonds, Kurds, Turks and Arabs. 
Politics, travel and research in north-eastern Iraq 1919- 
1925, London 1957, 195. (J.H. Kramers*) 
SARMAD [see muhammad sa c Id sarmad]. 
SARPUL-I DHUHAB ("bridgehead of Zohab"), 
a place on the way to the Zagros Mountains on 
the great Baghdad-Kirmanshah road, taking its name 
from the stone bridge of two arches over the river Al- 
wand, a tributary on the left bank of the Diyala. Sar- 
pul in the early 20th century consisted simply of a lit- 
tle fort {kur-khana = "arsenal") in which the governor 
of Zohab lived (the post was regularly filled by the 
chief of the tribe of Guran), a caravanserai, a garden 
of cypress and about 40 houses. The old town of 
Zohab, about 4 hours to the north, is now in ruins. To 
the east, behind the cliffs of Hazar-Djarib, lies the lit- 
tle canton of Beshiwe (Kurdish = "below") in a cor- 
ridor running round the foot of the Zagros giving ac- 
cess to the famous col of Pa-Tak, on the slope of which 
is the Sasanid edifice called Tak-i Girra. In the west, 
the heights of Mel-i Ya c kub separate the verdant plain 
of Sarpul from that of Kasr-i Shirln [q.v.]. Sarpul is 
the natural halting place for thousands of Persian 
pilgrims going to the c atabdt [q.v. in Suppl.] (Karbala 3 
and other Shi*-! sanctuaries). When the pilgrimage 
season is at its height (in autumn and winter), a hun- 
dred tents may be seen near the bridge. They belong 
to the Kurdish gipsy tribe of Suzmani (Fiyudj), the 
women of which are professional dancers and singers 
noted for their light morals. 

Sarpul corresponds to the site of the ancient 
Khalmanu of the Assyrians, Hulwan [q.v.] of the 
Arabs. The earlier name survived as the Kurdish 
name of the Alwand, i.e. Halawan. Traces of the old 
town are found mainly on the left bank (Paypul) 
where the land is level and beautiful. 

Sarpul is noted for its antiquities: (1) the bas-relief 
and Pahlavi inscription on the cliff on the right bank 
of the Alwand; (2) three steles on the cliffs of Hazar- 
Djarib (on the left bank) of which two are Sasanid 
(Parthian?) and the third represents Anu-Banini, king 
of the Lulubi; (3) two miles away, to the south of 
Hazar-Djarib, is an Achaemenid tomb cut out of the 
rock and venerated at the present day under the name 
of Dukkan-i Dawud (David's workshop) by the Ahl-i 
Hakk [q.v.], who have a cemetery at the foot of the 

Modern Sarpul-i Dhuhab is the chef-lieu of a bakhsh 
of the same name in the shahrastdn of Kasr-i Shirln in 
Kirmanshah province (ustdn) (lat. 33° 27'N., long. 
45° 25' E., alt. 534 m/1,750 feet). In ca. 1950 it had 
a population of around 2,000, comprising Shl'Is, Ahl- 
i Hakk and Sunms, a number swollen in winter how- 


:r by the influx of transhum; 
Bibliography: H. Rawl 

:s from their yaylakt 

son, in JRGS, ix 
(1839), 39; Ritter, Erdkunde, ix, Berlin 1840, 460; 
J.F. Jones, Memoirs in Selections from the records of the 
Bombay Government, N.S., xliii, 150; Cirikov, Putevoy 
Zhurnal, St. Petersburg 1875, 313 and passim; J. P. 
Ferrier, Voyages en Perse, Paris 1860, i, 29; J. de 
Morgan, Miss, sclent., ii, Etudes geogr. , Paris 1895, 
106, iv, Recherches archeol., part 1, Paris 1896, 149- 
71 (plates VII and XII give detailed sketches of the 
locality); E. Aubin, La Perse d'aujourd'hui, Paris 
1908, 348; Sarre-Herzfeld, Iramsche Felsreliefs, 
Berlin 1910, 61; Herzfeld, Am Tor von Asien, Berlin 
1920; Razmara (ed.), Farhang-i djughrafiya-yi Iran- 
zamin, v, 230; Sylvia A. Matheson, Persia, an arch- 
aeological guide 2 , London 1972, 135-6, 294. 

(V. Minorsky*) 
SARRADJ [see sardj]. 

SARRADJ, BANU 'l [see ibn al-sarradj, in 
Suppl.]. . 

AL-SARRADI, Abu Muhammad Dja c far b. 
Ahmad b. al-Husayn al-Kari 5 (d. 500/1106), known 
as DjaTar al-Sarradj, or al-Sarradj al-Baghdadi, a 
Hanball traditionist from Baghdad, noted for his 
poetry and author of the famous Masari' al-'ushshak. 
was born in Baghdad at the turn of the year 
417/February 1027. 

He took up the study of Islamic Tradition at a very 
early age and received an education in the kira^at 
[q.v.]. He actually taught this for several years, not 

belles lettres and his poetic skills. Thus al-Khatlb al- 
Baghdadl (d. 473/1071) edited a collection of his 
enunciations ifawpid), and specimens of his poetry 
are given by all of his biographers. Al-Sarradj is said 
to have been proud of the many authorised transmis- 
sions (riwaya [q.v.]) he had gathered, and he con- 
tinually collected — and transmitted — traditions of va- 
rious kinds. He travelled to Mecca, Egypt and, a 
number of times, the coastal towns of Syria, especially 
Tyre, and to Damascus where, according to Ibn 
c Asakir, his (late) transmission from the famous Ibn 
Shadhan (d. 426/1034) was appreciated. Among his 
students figures the well-known Abu Tahir al-Silafl 

Titles of two kinds of his writings are mentioned in 
the sources, one being the versifications of works on 
fikh and religous matters, as e.g. the K. al-Tanbih by 
Abu Ishak al-Shlrazi (d. 476/1083), a certain Mandsik 
al-hadjcgjmdzK. al-Mubtadd\ The al-Urdjuzafi nazd'ir 
al-Kur^an, mentioned by Brockelmann, seems to 
belong to this group. Other titles apparently concern 
works of a moralising adab type, such as his Hukm al- 
sibydn and Mandkib al-habash (al-suddn). 

His Masari' al-'ushshak, "The places (or: times, oc- 
casions) of slaughter of passionate lovers" (or, if 
masari'- is taken as the exceptional plural form of ran*, 
"Lovers who were slain [by love]"), agrees in charac- 
ter with this second type of writings. It is a collection 
of short narrations, often accompanied by verses, 
about the calamity of love, which is depicted as a 
chaste and, mostly, a fatal experience. The author 
only contributes some verses at the end of each section 
(^W), and meticulously quotes isnads. They show 
that these materials circulated among scholars of 
Islamic Tradition, and that they were taken, to a large 
extent attributed to the founders of adab literature, 
from earlier collections and, to a lesser degree, from 
Sufi teachings. His book, which served as a basic 
source for later works on this topic, was classified even 
by mediaeval authors as 'adjib "remarkable", and, in 

modern studies, has raised questions as to the 
underlying teaching on love and its connection with 
the Hanbalr school of thought. Neither the origin of 
heir presentation, which is void of any 

character of the work (Bell), but the edifying effect of 
(dhikr) and the impersonal perspective 
prevalent in most of the narrations (Vadet) den 

parable to earlier collections of edifying narrations or 
"reports", which were composed by representatives 
of the Ahl al-Sunna wa 'l-Djama'a, such as Ibn Abi '1- 
Dunya and others. 

Bibliography: There is no critical edition of the 
Masari'; most versions are reprints of the Beirut 
1958 ed.— Entries on al-Sarradj are found in most 
of the biographical literature of the period con- 
cerned: Ibn al-Djawzf, al-Muntazam, 18 vols., 
Beirut 1412/1992, xvii, 102-4; Yakut, Irshad, ed. 
Margoliouth, ii, Leiden 1909, 401-5; Ibn Manzur, 
Mukhtasar Ta^nkh Dimashk, vi, Damascus 
1404/1984, 52; Ibn al-Nadjdjar, al-Mustqfdd mm 
dhayl TaMkh Baghdad, intika^ Ibn al-Dimyatl, Beirut 
n.d., 93-5; Ibn Radjab, Dhayl c ala Tabakat al- 
' ' ' " ' - ■ -• i Cairo 1952, 

ed. M. Hami 

s of h 

= Brockelm, 

, I 2 , 

431, S I 594; J. -CI. Vadet, L 'esprit courtois en Orient 

dans les cinq premiers siecles de Vhegire, Paris 1968, 379- 

430; J.N. Bell, The Hanbalite teaching on love, diss. 

Princeton University 1971 (University Microfilms), 

245-58; S. Leder, Ibn ai-Gauzi und seine Kompilation 

wider die Leidenschaft, Beirut-Wiesbaden 1984, 98- 

101; German tr. of excerpts from the Masari', by R. 

Paret, in Fruharabische Liebesgeschichten. Ein Beitrag 

zur vergleichenden Literaturgeschichte, Sprache und 

Dichtung 40, Bern 1927. (S. Leder) 

al-SARRADJ, Abu Nasr c Abd Allah b. c Ali, Sufi 

author originally from Tus in Khurasan, who lived 

towards the middle of the 4th/ 10th century and died 

in his home town in 378/988. 

The bio-bibliographical and hagiographic literature 
( c Attar, Tadhkira, ed. Nicholson, ii, 182-3; al- 
Dhahabi, Ta\ikh al-lsldm, cited by Nicholson in The 
Kitdb al-Luma', p. Ill; Ibn Taghrlbirdl, Nudjum, iv, 
152; Djami, Nafahdl, no. 353; Ibn al- c Imad, Shadharat, 
iii, 91) gives hardly any precise information about his 
life and upbringing. We do, however, know the 
names of some of his teachers and sources of informa- 
tion e.g. amongst others, Dja'far al-Khuldi (d. in 
Baghdad 348/960) and Ahmad b. Muhamad b. Salim 
of Basra (d. 356/967), the leading light of the 
Salimiyya school. The fact that he frequented the 
company of this last and that he cites him does not, 
nevertheless, allow us to count him amongst the par- 
tisans of the Salimiyya [q.v.] (see Nicholson, op. cil. , 
pp. XI-XII); the lively discussion which he had with 
Ibn Salim on the validity of the shatahat attributed to 
Abu Yazid al-Bistami (Luma', ch. Tafsir al-shathiyyat) 
testifies all the same to his independence of mind. 
From the text of the K. al-Luma' it appears that al- 
Sarradj travelled widely, since he cites, on occasion, 
conversations which he apparently had not only in 
Persia but also in 'Irak, Syria and even Egypt. Above 
all, our sources mention the remarkable consequences 
of his spiritual elevation. Thus, invited during 
Ramadan to direct the education of the dervishes in 

dyya mosque at Baghdad, he is 

the Shur 

have touched for a month the food which w 

to his cell {Tadhkira, ii, 182; Nafahdt, no. 353; 

Hudjwin, KasJif al-mahdjub, ed. Zhukovski, 417). 

' Attar, and Djami following him, further report that 


al-Sarradj, carried off by a moment of ecstasy, plung- 
ed his face into a flaming brazier without suffering 
any pain or leaving any trace on his face; on the con- 
trary, it was completely radiant. But in fact, we know 
hardly anything further on his life and, in particular, 
have no information at all on his possible successors 
on the mystical path. As for his disciples, we know on- 
ly one name, that of Abu '1-Fadl of Sarakhs, the future 
master of Abu Sa c Id b. Abi '1-Khayr [q.v.] (Muham- 
mad b. al-Munawwar, Asrdr al-tawhid, ed. Dh. §afa, 
27; Nafahat, no. 354). This is a remarkable dearth of 
information about one who was called "the peacock of 
the poor" (tdwus al-fukard 7 ) and whose authority and 
competence were widely recognised. 

In fact, the personality of al-Sarradj is completely 
hidden behind the K. al-Luma' "Book of shafts of 
light", his main and probably only work, since, 
despite an affirmation of existence by DjamI, no other 
title attributed to him is known to us. This work in- 
volves a treatise of considerable value, both from the 
richness of its documentary information on the Sufism 
of the first Islamic centuries and also from the quality 
of the religious thought which informs it. Al-Sarradj 
presents there the bases of knowledge understood in a 
mystical sense, contrasted with the Islamic religious 
sciences known at the time. He details in it the main 
stages and states of the mystical path; underlines the 
importance of the revealed sacred text and the pro- 
phetic example by highlighting the ways of interpreta- 
tion followed by the Sufis; describes the customs (adab) 
of Sufis and cites the particularly significant texts of 
the great masters. The precision of his definitions in 
the technical lexicon of Sufism is most valuable. He 
also deals with the tangible and controversial aspects 
of the Sufi life, such as the status of miracles (kardmdt), 
the nature of ecstasy, the lawfulness of listening to 
music (samd' [q.v.]), the orthodoxy of the paradoxical 
utterances (shatahdt) attributed to certain mashdyikfi; 
and the doctrinal errors of several currents of thought 
claiming a connection with Sufism. Each chapter 
forms a little, autonomous treatise, in which the 
author cites abundantly hadiths and, especially, the 
dicta of the great masters of Sufism. His own points 
of view are not concealed at all but are fitted in fairly 
discreetly behind the teachings of the great figures in 
the tradition. 

The importance of the Luma' was appreciated as 
soon as it appeared. Subsequent authors, like al- 
Kushayn in his Risdla, found in it substantial bases of 
documentation, as did even Abu Hamid al-Qhazall, 
who drew upon it for several elements in his writings 
on the behaviour of Suffs (see esp. the Ihyd^, .Book ii, 
K. Adab al-sama' wa 'l-wadjd). The Luma'- also con- 
tributed to the legitirr.isation of Sufism as an Islamic 
science in its own right. Al-Sarradj showed himself 
quite firm about the essential point: true Sufis are not 
merely in complete conformity with Islamic or- 
thodoxy but they themselves make up its spiritual 
elite. It is not, then, a question of an apologia, in the 
strict sense of the term, or of a purely defensive 
justification of Sufism, but it goes beyond that to an 
argued and assured statement of the harmonious in- 
tegration of mysticism within the bosom of Muslim 
religious life. Being moderate and aiming at a consen- 
sus of opinion, al-Sarradj's language in the Luma' thus 
forms a particularly clear and vivid example of the 
conception which the Sufis had of themselves towards 
the middle of the 4th/10th century. 

Bibliography: The Kitdb al-Luma*..., ed. with 
summary and glossary, by R.A. Nicholson, GMS, 
Leiden-London 1914; A.J. Arberry (ed.), Pages of 
the Kitdb al-Luma'- of Abu Nasr al-Sarrdj, London 1947 

(gathers together the lacunae of the Nicholson ed. 
on the basis of a Bankipore ms., with a memoir, 
preface and notes); Schlaglichter iiber das Sufitum, the 
K. al-Luma', introd., tr. and comm. by R. 
Gramlich, Wiesbaden 1990 (a valuable work, in- 
cluding a correction of Nicholson's text from two 
supplementary mss. as well as exhaustive references 
throughout the text to the Arabic and Persian, Sufi 
and non-Sufi, bio-bibliographical sources). 

(P. Lory) 
SARRUF, Ya c kub, a personage of the revival 
of Arabic culture and literature or nahda {q.v.} 
(b. al-Hadath, 18 July 1852, d. Cairo, 9 July 1927). 
Of Maronite origin, he was part of the first wave of 
graduates from the Syrian Protestant College in 1870 
and taught for two years in American missionary 
schools. Little is known about his conversion to Pro- 
testantism. From 1876 onwards, together with Faris 
Nimr [q. v. ] , his name was, for half a century, attached 
to the journal al-Muktataf, which published 121 
volumes over its life-span of 75 years. In order to 
avoid official displeasure, he emigrated on March 
1885 to Egypt, with his journal. Unceasingly remain- 
ing as close as possible to contemporary issues, he 
translated and popularised the latest scientific 
discoveries, defended freemasonry, and clashed with 
the Jesuits of the journal al-Mashrik and the reformists 
of the journal al-Mandr [q.v.], since he was only in- 
terested in religion (as a Christian within an Islamic 
society) or politics (thus he did not take up a position 
against the British occupation of Egypt) from the 
scientific viewpoint. He championed Darwinian 
evolution, even though his lack of a specialist scientific 
training did not allow him fully to understand the 
philosophical bases of that doctrine. As well as 
numerous translations, he published three novels as 
separate volumes, Fatal Misr, Amir Lubndn and Fatal al- 
Fayyum), as well as biographical collections and works 
on natural history (Fusul fi 'l-taMkh al-tabCi min 
mamlakatay al-hayawdn wa 'l-nabdt) and astronomy 
(BasdHt Him al-falak wa-suwar al-samd 7 ). As a journalist 
pre-occupied with science, he not only invented words 
to translate new English and French concepts into 
Arabic but also endeavoured to express himself in his 
own writings in a clear language. He proposed a 
theory of his own practice in this regard. His work 
shows the limitations of a person who acted as a chan- 
nel of transmission for a modern field of knowledge 
without taking part in its elaboration. 

Bibliography: Brockelmann, S III, 215-17; Sa'd 
Abu Diyya, al-A/kdr al-mdsuniyya fi madjallat al- 
Muktafaf, in al-Madjalla al-Ta^rikhiyya al-'Arabiyya li 
'l-Dirdsdt al-'Uthmdniyya, nos. 1-2 (Jan. 1990), 9-19; 
Daghir, Masddir, ii, 540-8; Nadia Farag, al-Muqtataf 
1876-1900; a study of the influence of Victorian thought on 
modern Arabic thought, diss. Oxford Univ. 1969, un- 
publ.; Fihris al-Muktataf Beirut 1967, 3 vols.; 
Kahhala, xiii, 353-4; Anls al-MakdisI, al-Funun al- 
adabiyya wa-a'ldmuhafi 'l-nahda al-'arabiyya al-haditha, 
Beirut 1963, 239-57; Fu'ad Sarruf, in Ruwwdd In- 
djiliyydn, 103-27; Ph. TarrazT, Ta\tkh al-sahdfa, ii, 
124-9; c Abd Allah al- c Umar, al-Madjalla al- 
thakafiyya wa 'l-tahwirdt al-mu'dsira, Kuwait 1984, 9- 
68_; Zirikll, ix, 266. (J. Fontaine) 

SART, a term found in the history and 
ethnography of the Persian and Central Asian 
worlds. Originally an old Turkic word for "mer- 
chant", it occurs with this meaning in the 11th- 
century sources, such as Mahmud al-Kashghari's en- 
cyclopaedic dictionary Diwdn lughdt al-Turk (Compen- 
dium of the Turkic dialects, tr. R. Dankoff and J. Kelly, 
3 vols., Cambridge, Mass., 1982-5, i, 269), and the 

Karakhanid mirror for princes, Kutadgu bilig by Yusuf 
Khass Hadjib(ed. R.R. Aral, Istanbul 1947, i, 571). 
For references to other editions of both works, see 
Drevnetyurkskiy slovar' , Leningrad 1969, 490). 

Evidently a loan-word from the Sanskrit for 
"caravan leader" (sdrthavdha), it is attested in the 
Uighur translation of the Saddharmapundarika sutra as 
sartbav, which is explained as "head of the merchants" 
(satighdjilar ulught) (Kuansi im pusar, ed. and tr. 
§. Tekin, 1960, repr. Ankara 1993, 11), and in the 
Manichaean and Buddhist Uighur texts from Turfan 
(W. Bang and A. von Gabain, Tiirkische Turfan-Texte 
II and Tiirkische Turfan-Texte V, in Sprachwissenschaft- 
liche Ergebnisse der deutschen Turfan-Forschung, 3 vols., 
repr. Leipzig 1972-85, ii, 34 and n. 16, 120 and n. 
B57, the latter in the form sarlavaghi, which was also 
a designation for the Bodhisattva. See also Bartol'd, 
ii/1, 162, 196-7, 253). It may have come into Uighur 
through Sogdian (G. Clauson, An etymological dictionary 
of pre-thirteenth-century Turkish, Oxford 1972, 846), 

over the trade with the Turkic peoples from the In- 
dians (Bartol'd, ii/1, 460). 

When the western Iranians secured control of the 
trade with the nomadic peoples, the Turks and 
Mongols applied the term sari to them in the same 
sense as tddjik [q.v.\ (Bartol'd, ii/2, 304). Because Ira- 
nians were a sedentary Muslim people, the term also 
designated all sedentary Muslims, irrespective of 
language or ethnicity (Fragner, 21). Thus, the 
Mongols referred to Arslan Khan, the prince of the 
Muslim Turkic Karluks, as sartdktai (tai being the 
masculine ending), which Rashid al-Din explained as 
meaning Tadjik (Djdmi c al-lawdrtkh, i/1, ed. A. A. 
RomaskeviJ et alii, Moscow 1965, 351). The sartdktai 
were therefore viewed by the Mongols not just as mer- 
chants, but as bearers of Perso-Islamic civilisation. 
This is confirmed by the fact that the word sartdul, 
which is obviously derived from the same root (e.g. 
ibid., ii, ed. E. Blochet, Leiden 1911, 541), occurs in 
Ibn Muhanna's late 14th-century polyglot glossary 
with the meaning al-muslimun (P. Melioranskiy, Arab 
filolog o mongol'skomyazike, in ZVOIRAO, xv [1904], 75 
and 136). It was this term that was applied by the 
Mongols to the Kh w arazmshahs and their subjects 
(Bartol'd ii/1, 461, ii/2, 310-11). 

In the post-Mongol period in Central Asia, the 
term sdrt was used synonymously with tddjik to 
designate the Persian-speaking sedentary/urban 
population in contrast to tiirk, which was used for the 
Turkic-speaking nomadic or semi-nomadic popula- 
tion. The 15th-century Caghatay Turkish author Mir 
c Ali-Shir Nawa'I [q.v.] regularly referred to the Ira- 
nian people as sart ulusi and to their language as sari 
tili, the latter as a synonym for farsi (Persian). He also 
stated that in Khurasan the Sarts did not speak 
Turkish and if they did, it was clearly recognisable 
that they were Sarts (Muhdkamat al-lughatayn, ed. and 
tr. R. Devereux, Leiden 1966, 6. See also Abuska lugati 
veya Cagatay sozlugii, ed. B. Atalay, Ankara 1970, 264- 
5, for other instances of its use by Nawa'i in the 
Madjdlis al-nafdHs, etc.; also Pavet de Courteille, Dic- 
tionnaire turc-oriental, Paris 1870, 334). In his descrip- 
tion of Farghana, Babur referred to the population of 
Andidjan as tiirk, while he said that the inhabitants of 
Marghfnan and Asfara were sarts,, adding that the lat- 
ter were Persian-speaking (fdrsi-guy) (The Bdbar-ndma, 
facs. ed. A.S. Beveridge, 1905, repr. London 1971, 
fols. 2b, 3b; Bdbumdma, Chaghatay Turkish text with 
Abdul-Rahim Khankhanan's Persian translation, Turkish 
transcription, Persian edition and English translation 
by W.M. Thackston Jr., Pt. 1, [Cambridge, Mass.] 

1993, 6-7 (note that Khankhanan renders sdrt as 
tddjik)). Elsewhere, he pointed out that the urban in- 
habitants of Kabul were sdrts (fols. 131 a-b; ed. and tr. 
Thackston, Pt. 2, [Cambridge, Mass.] 1993, 270). 

Following their conquest of Central Asia in the 1 6th 
century, the nomadic Uzbeks clearly distinguished be- 
tween themselves and their sedentary subjects whom 
they called sarts, irrespective of linguistic or ethnic 
considerations. The distinction between Uzbek and 
Sart (or Tadjik) was now felt more keenly than the 
older distinction between Tiirk and Sart, probably 
because some of the pre-Uzbek Turkic (Caghatay) 
tribes had already made the transition to sedentarism. 
Abu '1-GhazI. the 17th-century Caghatay Turkish 
author of the Shadjara-yi Turk, frequently employed 
the expression "Uzbeks and Sarts" when referring to 
the entire population of Kh w arazm (Histoire des 
mongols et des tatares par Aboul-Ghdzi Behddour Khan, ed. 
and tr. P.I. Desmaisons, 1871-4, repr. Amsterdam 
1970, 231, 256). The same usage survived to the 19th 
century, as attested in the works of historians of 
Khlwa. such as Mu'nis, Agahl and Thana 3 !, although 
it also took on political significance (Bregel, 123-5; 
Bartol'd, v, 223), as well as of Khokand (see T.K. 
Beisembiev, "Ta'nkh-i shakhrukhi" kak istonceskiy 
istocnik, Alma-Ata 1987, 93). 

The introduction of a critical mass of Turkic and 
Turkicised Mongolian nomads into Transoxiana as a 
consequence of the Uzbek invasions strengthened the 
general trend toward Turkicisation of the indigenous 
Iranian population. The degree to which Turkicisa- 
tion and Uzbek sedentarisation had progressed by the 
19th century is vividly indicated by the fact that the 
name Sart, which had first been applied to sedentary 
Iranians and then to both Persian and Turkic- 
speaking urban dwellers, began to be used chiefly for 
Turkic-speaking or bilingual town-dwellers, while the 
term Tadjik, which had earlier been synonymous with 
Sart, was reserved for Persian-speakers only 
(Subtelny, 49-50; Fragner, 22). Moreover, it was now 
Kazaks (or Turkmens in the case of Kh w arazm) who 
as nomads were contrasted with the Sarts as urban 
dwellers and agriculturalists in Bukhara and Khokand 
(Bartol'd, ii/1, 462, v, 223; Bregel, 123). 

Although 19th- and early 20th-century European 
and Russian scholars adopted the use of the term Sart 
for urban Turkic-speakers, who were regarded as the 
"native inhabitants" of Central Asia (Subtelny, 50; 
Bartol'd, ii/1, 462, ii/2, 303-5; L. Budagov, Srav- 
nitel'niy slovar' turetsko-tatarskikh nareciy, St. Petersburg 
1869, i, 612), Bartol'd himself remained sceptical 
about this use, since he found no direct proof for it in 
the historical sources (ii/1, 462). No serious attempt 
was ever made to determine the dialectal differences 
between Sarts and Uzbeks either in Bukhara or in 
Khlwa. although they certainly existed and probably 
warranted the treatment of Sarts as a separate ethnic 
group (Bregel, 148; Bartol'd, ii/2, 303 ff.). However, 
the close symbiotic relationship that had existed for 
centuries between the (nomadic) Turkic and (seden- 
tary) Iranian peoples in Central Asia, resulting in 
mutual linguistic and ethnogenetic influences, made it 
extremely difficult, if not impossible, to sort out the 
differences between the two (Subtelny, 44 ff.). 

During the 1920s, in keeping with Soviet na- 
tionalities policies in Central Asia, which aimed at 
creating separate national republics based primarily 
on ethnolinguistic criteria, the name Sart was banish- 
ed from use and the ethnic designation Uzbek 
substituted for it on the grounds that Sart had never 
been an ethnic term and that it was insulting to the in- 
digenous population on account of its pejorative 


popular etymology "yellow dog" (sari it), which 
reflected the traditionally contemptuous attitude of 
the nomads toward sedentary peoples (Subtelny, 48, 
51; Bartol'd, ii/2, 303; for other popular etymologies 
see Budagov, 612), although Bartol'd argued that 
there was nothing pejorative about the term and it had 
in fact been used by the Sarts as a self-designation 
(ii/1, 462, ii/2, 314; B. Kh. Karmisheva, Ocerkt et- 
niceskoy istorii yuinikh rayonov Tadiikistana i Uzbekistana, 
Moscow 1976, 147). Soviet historiography never dealt 
adequately with the term or with the Sarts as an 
ethnico-linguistic group (for an overview, see Bregel, 
121-2), although rich ethnographic material exists for 
its study (see e.g. the bibliography in E.A. Voznesen- 
skaya and A.B. Piotrovskiy, Material! dlyabibliografii po 
antropologii i etnografii Kazakstana i sredneaziatskikh 
respublik. Trudl Komissii po izuceniyu plemennogo soslava 
naseleniya SSSR i sopredel'ntkh stran, xiv, Leningrad 
1927, 181-99). 

Bibliography: V.V. Bartol'd, Sodneniya, 10 
vols., Moscow 1963-77, esp. prepodavanii tuzem- 
nlkh nareciy v Samarkande, ii/2, 303-5; Eshce o dove 
"sort", ii/2, 310-14; Sart, ii/2, 527-9 (tr. of the cor- 
rected version of his article in EP ); Yu. Bregel, The 
Sarts in the khanate oj Khiva, in Journal of Asian history, 
xii/2 (1978), 120-51; B.G. Fragner, Probleme der Na- 
lionswerdung der Usbeken und Tadschiken, in Die 
Muslime in der Sowjetunion und injugoslawien, ed. A. 
Kappeler a alii., Cologne 1989, 19-34; M.E. 
Subtelny, The symbiosis of Turk and Tajik, in Central 
Asia in historical perspective , ed. B.F. Manz, Boulder, 
Colo. 1994, 44-60. 

(W. Barthold-[M.E. Subtelny]) 
SART, the form of the name in Ottoman Turkish 
of the small village in Lydia in Asia Minor, the 
ancient Sardes (<xE Sdp8ti; of the classical authors, 
which makes Sam! Bey write Sard), capital of the Ly- 
dian kingdom, situated on the eastern bank of the Sart 
Cay (Paktolos) a little southward to the spot where this 
river joins the Gediz Cay (Hermos). Although in the 
later Byzantine period Sardes had lost much of its 
former importance (as a metropolitan see) and been 
outflanked by Magnesia (Turkish Maghnlsa [q.v.]) 
and Philadelphia (Ala Shehir [q.v.]), it still was one of 
the larger towns, when the Saldjuk Turks, in the 
5th/llth century, made incursions into the Hermus 
valley. At the time, they were expelled by the Byzan- 
tine general Philocales (1118). At the end of the 
7th/13th century, Sardes had been for some time 
under a combined Greek and Turkish domination, 
until the Greeks were able to drive away the Turks a 
second time (Pachymeres, ed. Niebuhr, Bonn 1835, 
ii, 403). In the beginning of the 8th/14th century the 
citadel was surrendered to one of the Saldjuk amirs, 
and the town probably belonged during the remainder 
of that century to the territory of the Sarukhan [q. v. ] 
dynasty, whose capital was Maghnlsa. So when in 
792/1390 the Ottoman sultan Bayezid I, after the con- 
quest of the then Greek town Philadelphia, took 
possession of the Sarukhan country, Sardes was 
equally incorporated in his empire (Anonymus Giese, 
Breslau 1922, 28; 'Ashik-pasha-zade, Istanbul 1333, 
65). After the battle of Ankara, when Timur marched 
against Izmir (805/1402), Sardes and its citadel were 
probably destroyed and never recovered again. 

The modern settlement of Sart (lat. 38° 28' N., 
long. 28° 03' E., alt. 102 m/335 ft.) lies between the 
Sart Cay and the citadel hill. This hill is a long, nar- 
row counterfort, 200 m/656 ft. in height, belonging to 
Mount Tmolus (now Mahmud Dag in the south (a 
topographical sketch of the site in Curtius, Beitrdge zur 
Geschichte und Topographie Kleinasiens, in Abh. Pr. Ak. 
W. [1872], Plate V 2 ). East of the ridge is a small mill 

brook called Tabak gay; north of the town it joins the 
Sart gay, which is united with the Gediz gay about 
6 km/9 14 miles to the north of the acropolis hill. At 
the other side of the Gediz gay is situated the big 
necropolis of Sardes, a large plain of mounds called 
Bin Bir Tepe. North of this plain is the Mermere Lake, 
the ancient Lake of Gyges. The railway from Izmir to 
Alasehir runs along the southern Gediz gay bank and 
has a station at Sart. Administratively, Sart now 
comes within the nahiye of Salihh in the it or province 
of Manisa. 

Excavations at the classical, Roman and Byzantine 
site of Sardis have been undertaken, especially by 
American archaeologists since the early part of the 
20th century. In recent decades, these have been 
associated with G.M.A. Hanfmann. See, in par- 
ticular, his Sardis und Lydien, in Abh. Ak. Wiss. zu 
Mainz, geistes- un sozial wiss. Kl.,Jhg. 1960, 499-536; 
idem (ed.), Archaeological exploration oj Sardis (1958- 
1975). Sardis from prehistoric to Roman limes, Cambridge, 
Mass. 1983. See also PW, 1.A.2, cols. 2475-8 s.v. 
Sardeis, and Der kleine Pauly, iv, cols. 1551-2. 

Bibliography: Hadjdji Khalifa, Djihdnnumd, 
Istanbul 1145/1732-3, 636; Sami, Kdmus al-aHam, 
iv, 2477; E. Banse, Die TurkeP, Brunswick 1919, 
119, 132-4; von Hammer, GOR, i, 70; V. Cuinet, 
La Turquie d'Asie, Paris 1894, iii, 532, 533, 565; Ad- 
miralty Handbook. Turkey, London 1942-3, ii, 317-18, 
320. _ G-H. Kramers*) 

SARUDI, a town in Diyar Mudar [q.v.] on the 
most southerly of the three roads from Biredjik [q.v.] 
to Urfa [see al-ruhA] in 36° 58' N. lat. and 38° 27' 
E. long. As the name of the town is also that of the 


, the z 

hemusia and Batnae is disputed; see Bibl. On account 
of the fertility of the district in which the town is 
situated, and its central position between the 
Euphrates on the one side, and Urfa and Harran 
[q.v.], from each of which it is about a day's journey 
distant, on the other, the traffic through it brought it 
a certain degree of prosperity, especially as it was also 
important as a post-station between al-Rakka and 
Sumaysat. According to Ibn Khurradadhbih, it was 
20farsakhs from the former town and 13 from the lat- 
ter. The principal occupation was settled by the 
natural suitability of the soil or growing fruit and the 
vine, as all the geographers tell us. Within the town 
itself Ibn Djubayr (late 6th/ 12th century) found or- 
chards and running water. 

Sarudj— an ancient centre of Syriac Christianity, 
and the birthplace of Jacob of Sarudj (d. 521)— was 
captured, with the rest of al-Djazira, by the Arab 
commander c Iyad b. Ghanm in 18/639. Three cen- 
turies later it came within the possessions of the Ham- 
danids [q.v.] of Mawsil and Aleppo, and then of the 
resurgent Byzantines. Towards the end of the 
5th/ 11th century, Diyar Mudar was controlled by the 
Turkmen family of the Artukids [q.v.], proteges of the 
Saldjuks, and Sukman b. Artuk made Sarudj the cen- 
tre of his lands there until the Crusaders under 
Baldwin of Bouillon appeared at nearby Urfa or 
Edessa in 1098. For nearly fifty years, the town— at 
this time known as the seat of a Jacobite bishopric- 
formed part of the Frankish County of Edessa. The 
Franks held Sarudj, in the face of attacks by the Ar- 
tukid Il-GhazI and the Turkish governor of Mawsil, 
Mawdud, until in Radjab 539/January 1145 it fell to 
c Imad al-Din Zangi [q.v.] and reverted to Muslim 
control. It was as a native of the town, expelled thence 
by the Christians, that al-Harln [q.v.] made the hero 
of his makdmat [q.v.], Abu Zayd al-SarudjI; the 

By the time of Abu '1-Fida' (8th/14th century), 


Sarudj was already in ruins. Nineteenth-century 
travellers described it much as did the mediaeval 
geographers, except that it appeared smaller to them. 
Sachau (see Bibl.) actually spoke of the village of 
Sarudj; in late Ottoman times it was the residence of 
a kPim-makam. 

In modern times, the town was occupied by British 
and then French troops in 1920 after the First World 
War, but with the delimitation of the frontier, came 
within Turkey, so that Siiriic is now just north of the 
Syrian frontier. Administratively, it lies within the 
province or il of Diyarbakir and is the chef-lieu of a 
county or ike also called Suriic; in I960 it had a 
population of 6,800. 

Bibliography: Fraenkel, arts. Anthemusia and 
Batnai in Pauly-Wissowa; K. Regling, Zur histor. 
Geographic d. mesopol. Parallelogramms, in Klio, i 
(1902), 443-76; H. and R. Kiepert, Formae orbis ant- 
iqui, 1910, part v, 5b; Le Strange, Lands of the 
Eastern Caliphate, 108; Istakhn, 78; Ibn Hawkal, ed. 
Kramers, 230, tr. Kramers-Wiet, 224; Ibn 
Djubayr, Rihla, ed. Wright-de Goeje, 248; Yakut, 
Mu c djam, index s.v.; Abu '1-Fida 3 , Takwim al- 
bulddn, ed. Reinaud and de Slane, 233-4, 276; tr. ii, 
12-13, 52; Ibn Khurradadhbih, 73, 97, 216; Ibn al- 
Athir, index s.v.; R. Pococke, Description of the East, 
Book ii, ch. xviii, German tr. E. von Windheim, 
Erlangen 1754, ii, 238; W.F. Ainsworth, Travels and 
researches, London 1842, ii, 102-3; E. Sachau, Reise 
in Syrien und Mesopotamien, Leipzig 1883, 180-1; M. 
von Oppenheim, Vom Mktelmeer zum Pers. Golf, 
1900, ii, 3, and the map, western sheet; Caetani, 
Annali dellTslam, Milan 1911, iv, 32-48; E. 
Honigmann, Studien zur Notitia Antiochena, in Byz. 
Zeitschrift, xxv (1925), 73, 77-8; CI. Cahen, La Syrie 
du nord, 113, 180, 187, 219, 230, 263-4, 291; M. 
Canard, Histoire de la dynastic des H'amdanides, 92-3; 
F. Isiltan, Urfa bblgesi tarihi, Istanbul 1960, index; 
J.B. Segal, Edessa, 'The Blessed City', Oxford 1970, 
index; IA, art. Suriic. 

(M. Plessner-[C.E. Bosworth]) 
SARUKHAN, the name of a Turkish amirate, 
which appeared after the collapse of the Saldjuk state 
of Rum [see saldjukids. III. 5]. 

It was probably named after its founder, Sarukhan 
the son of Alpagi, and it was situated in the region 
roughly coinciding with the ancient Lydia, a territory 
yielding a rich agricultural production (grapes, figs 
and especially cereals). Its capital was the ancient city 
of Magnesia on the Sipylos, called by the Turks 
Maghnisa [q.v.] or Manisa, which, after having ac- 
quired special importance under the emperors of 
Nicaea, was conquered by Sarukhan Beg ca. 1313. In 
the beginning, the Turks of Sarukhan extended their 
territory by fighting against the Byzantines and, ac- 
cording to the Ottoman chronicler Enweri, they join- 
ed the Turks of Aydin [q.v.\ in naval raids directed 
against the Christian territories of the Aegean regions. 

esult of th 


slave market was created in Maghnisa. Nevertheless, 
the dangerous Genoese presence in their vicinity, i.e. 
in the alum-producing town of Phocaea as well as on 
the island of Chios [see sakiz], brought them into a 
rapprochement with the Byzantines. When the 
Emperor Andronicus III Palaeologus expelled the 
Genoese from Chios and compelled those of Phocaea 
to recognise his suzerainty, he concluded a treaty with 
Sarukhan (1329). Later, the amir, deeply vexed by 
the Genoese of Phocaea since his son Siileyman and 
twenty-four sons of Turkish notables were being held 
as hostages by them, offered military aid to the 
emperor against the Genoese governor of Phocaea, 

who rebelled and tried to conquer Lesbos; a new trea- 
ty was concluded (1335). During the Byzantine civil 
war (1341-7), Suleyman Sarukhan-oghlu followed 
Uraur Aydin-oghlu, the notorious ally of the 
pretender John Cantacuzenus, and fought on his side 
in Thrace, where he died after an illness (1345). In the 
light of a Genoese document, Sarukhan also died after 
1348 and was succeeded by his son Fakhr al-Din Ilyas. 
Around the same time, the amirate acquired a com- 
mon frontier with the Ottomans, who annexed the 
amirate of Karasi [q.v.]. 

An important part of the fleet of Sarukhan was 
destroyed by the Crusaders near Imbros during one of 
the usual raids organised in collaboration with Uraur 
Aydin-oghlu and directed against the Christian ter- 
ritories (1347). The exact year of Ilyas' death is not 
known. In 1357 the amirate was governed by his son 
Ishak Celebi, who was closely connected with the der- 
vish order of the Mewlesis or Mawlawiyya [q.v.]. It 
was probably Ishak Celebi who established relations 
with the Byzantine emperor John V Palaeologus when 
the latter sailed to Phocaea in 1358 in order to liberate 
the son of the Ottoman sultan Orkhan, Khalll, who 
was being held as prisoner by the Genoese. The 
emperor concluded a new treaty with the Sarukhan- 
oghlu and, furthermore, he took the amir's children 
as hostages to Constantinople. Ishak Celebi probably 
died in 1388 and was succeeded by his son Khidir 
Shah, in whose days the amirate was annexed by 
Bayezid I [q.v.] (1390). Khidir Shah was sent to Bursa 
where he died. 

Immediately after the battle of Ankara between 
TTmur and the Ottomans (804/1402), the amirate was 
restored by TTmur to Orkhan Sarukhan-oghlu, who 
had followed him in his Anatolian campaign. It was 
taken by the Ottomans again in 1410 to become one 
of the important sandjaks of their state, while 
Maghnisa became the residence of one of the heirs to 
the throne. Nomads of Sarukhan were transported by 
the Ottomans to Rumelia, and several soldiers were 
granted there timdrs (the SarOkhanlu). 

In the territory of Sarukhan and, especially, in 
Ma gh nisa, besides the important monuments erected 
by the Sarukhan-o gh ullari. new ones were added by 
the Ottomans to honour the residence of the Ottoman 
princes there; among them is the Muradiyye mosque 
(after Murad III), which was built by the great ar- 
chitect Sinan [q.v.]. 

Bibliography: H. Acun, Manisa'dakt turbe 
mimansi, in Belleten, xlix (1985), 479-501; M.M. 
Aktepe, XIV. ve XV. asirlarda Rumeli'nin Turkler 
taraftndan iskdmna dair, in Turkiyat Mecmuasi, x 
(1953), 300-2; 6.L. Barkan, Osmanh Imparatorlugun- 
da bir iskan ve kolonizasyon metodu olarak surgunler, in 
iklisat Fakultesi Mecmuasi, xv (1954), 215-16; F.M. 
Emecen, XVI. asirda Manisa kazasi, Ankara 1989; 
Kate Fleet, The treaty of 1387 between Murad I and the 
Genoese, in BSOAS, lvi (1993), 13-33; Irene 
Melikoff-Sayar, Le Destan d'Umur Pacha, Paris 1954; 
M.C. Ulucay, Saruhan ogullan ve eserlerine dair 
vesikalar (773H-1220H), Istanbul 1940; idem, in IA, 
art. Saruhan-ogullari; E.A. Zachariadou, Trade and 
crusade, Venetian Crete and the emirates of Menteshe and 
Aydin (1300-1415), Venice 1983; K. Zhukov, Ot- 
toman, Sarukhanid and Karasid coinage of the Hermitage 
collection, in The Ottoman emirate, 1300-1389, in 
Elizabeth Zachariadou, Halcyon days in Crete I. A 
symposium held in Rethymon 11-13 January 1991, 
Rethymon 1993, 237-42. 





1,597 m/5,238 ft.), some 80 km/50 miles to the 
southeast of Shlraz on the road to Nayriz [q.v.]. It 
seems to be identical with the Khawristan of the early 
Arab geographers, but first appears under the name 
Sarwistan ("place of cypresses") in al-Mukaddasi at 
the end of the 4th/10th century. 

It is notable for the tomb and shrine of a local saint, 
Shaykh Yusuf Sarwistanl, dated by its inscription to 
682/1283, and for a nearby mysterious building 
situated on the Shiraz-Fasa road 1 2 km/7 miles south 
of Sarwistan, first noted in 1810 by the English 
traveller Wm. Ouseley and described some 30 years 
later by Flandin and Coste. This has traditionally 
been taken by western scholarship as a Sasanid 
palace, building on its vague resemblance to the 
Emperor Ardashir's palace at Firuzabad [q. v. ] and on 
the fact that al-Tabarl, i } 870-1, tr. Noldeke, 
Geschichter der Perser und Amber, 111-12, says that 
Bahram V Gur's minister Mihr Narse laid out 
gardens near Firuzabad, including one with 12,000 
cypresses (sarw). Recently, however, a detailed ar- 
chitectural study by Lionel Bier has shown that it is 
more likely to be post-Islamic and is probably a 
Zoroastrian fire-temple erected in the last flowering of 
that faith in Fars ca. A.D. 750-950, though an origin 
in the Buyid period is not excluded. 

The modern town of Sarwistan is the chef-lieu of a 
ba khsh in the shahraslan of Shlraz. In ca. 1950 it had a 
population of around 4,000, but this has risen con- 
siderably in recent decades. 

Bibliography: Scl warz, Iran, 73-4; Le Strange, 
Lands, 252; G.N. Curzon, Persia and the Persian ques- 
tion, London 1892, ii, 111 n. 1; Razmara, Farhang-i 
djughrafiyd-yi Irdnzamm, vii, 129-30; Sylvia A. 
Matheson, Persia, an archaeological guide, London 
1976, 258-9; L. Bier, Sarvistan, a study in early Iranian 
architecture, Pennsylvania State Univ. Park and Lon- 
don 1986, with all the earlier bibl. given. 

(C.E. Bosworth) 
SASAN, Banu, the blanket designation in 
mediaeval Islamic literature for the practitioners of 
begging, swindling, confidence tricks, the 
displaying of disfiguring diseases, mutilated 
limbs, etc., so that sasanl has often become a general 
term in both Arabic and Persian for "beggar, 
trickster' ' . HadjdjI Khalifa uses sasani in the sense of 
"pertaining to magic or slight-of-hand", with the Him 
al-hiyal al-sasaniyya denoting "the science of artifices 
and trickery". 

In his treatise warning the general public against 
trickery in all forms, al-Mukhtar min kashf al-asrar, the 
early 7th/13th Syrian author c Abd al-Rahlm al- 
Djawbari [q.v. in Suppl.] enumerates the whole 
gamut of the members of the Banu Sasan: 
mountebanks, dervishes, gypsies (Zutt [q.v.]), those 
exhibiting bodily afflictions and feigning disease, 
those who teach tricks to animals and birds, and 
popular tellers of edifying tales and preachers (wu"az 
[see kass]), regarded as the highest class amongst the 
Sasanls; to these last, those who prey upon the 
populace's religious fears and scruples, al-Djawban 
devotes a whole chapter (see Bosworth, The mediaeval 
Islamic underworld, i, 24). 

Other material on the Banu Sasan can be culled 
from the rather exiguous sources in Arabic literature 
which deal with low life, including various adab works 
of al-Djahiz, Ibrahim al-Bayhakl's K. al-Mahasin wa 
'l-masawi and al-Raghib al-Isfahanl's Muhadarat al- 
udaba 1 , as well as those of the makama genre [q. v. ] and 
of the shadow plays [see ibn daniyal and khayal a'l- 
zill] . The present writer has studied in detail this ma- 
terial and also two long poems written substantially in 

the beggars' jargon (munaghat, lughat Bam Sasan) com- 
posed by the 4th/10th century traveller and litterateur 
Abu Dulaf al-Khazradji and by the 8th/14th century 
poet Saff al-DIn al-Hilll [q. vv. ] , in which many of their 
tricks and scabrous practices are delineated (The 
mediaeval Islamic underworld, i. The Banu Sasan in Arabic 
life and lore, ii. The Arabic jargon texts. The qaslda 
sasaniyyas of Abu Dulaf and Soft d-Din, Leiden 1976). 
The origin of the name Banu Sasan is shrouded in 
mystery. The sources give several explanations, some 
of them implausible. A mythical Shaykh Sasan is 
often mentioned as the founder of the fraternity of 
crooks and beggars. One oft-repeated story found in 
the sources, from the time of the Persian author and 
translator from Pahlavi into Arabic, Ibn al-Mukaffa c 
[q.v.], is that Sasan was the son of the legendary Per- 
sian emperor of the Kayanid line, Bahman b. Isfan- 
diyar, but was displaced from the succession to his 
father, hence took to a wandering live amongst, as 
some sources state, the nomadic Kurds (a people 
notorious in mediaeval Islamic times for banditry and 
violent ways of behaviour). Other stories hold that the 
Persians as a nation took to mendicancy after the 
Arab conquest of the lst/7th century, and aroused pi- 
ty and commiseration by claiming to be scions of the 
dispossessed Sasanid royal house. The process 
whereby the name of a fallen dynasty is applied 
ironically or satirically to a later group seems 
psychologically possible; but some modern authors 
have further suggested etymologies for sasan from 
Sanskrit or Persian (see Bosworth, op. cil. , i, 22-3). 
Bibliography: Given in the article, and see also 
that to al-djawbari; it should be noted that the 
critical text of the Kashf al-asrar promised by the 
author of that article has regrettably not yet ap- 
peared. (C.E. Bosworth) 
SASANIDS, a pre-Islamic Persian dynasty 
that ruled a large part of western Asia from A.D. 224 
until 651. In Arabic and modern usage, the dynastic 
name is derived from Sasan, who is mentioned as a 
"lord" in the inscription of Shapur I [q.v.] on the 
Ka c ba of Zoroaster (SKZ). The inscription ofNarseh 
at Paikuli also refers to the royal clan of Sasanagan. 
Theophoric names in the Parthian period suggest that 
Sasan may have been a minor deity or perhaps a 
deified ancestor. According to the late Sasanid Kar- 
ndmag, Sasan was the son-in-law of Papak, who gave 
him his daughter in marriage because Sasan was 
descended from Dara, the last Achaemenid king. Ar- 
dashir I [q.v.] was their child, although he is after- 
wards regarded as the son of Papak. In the account of 
al-Tabarl (i, 814), derived from the late Sasanid 
Xwaday-namag, Sasan was custodian of the fire temple 
of Anahita at Istakhr [q.v.], married to a member of 
the Bazrangid ruling family in Fars [q.v.], and the 
father of Papak. 

The rise of this family to power was aided by Par- 
thian (Arsacid) preoccupation with their own civil war 
and conflict with the Romans. In about 205-6 Papak 
overthrew the last Bazrangid ruler of Fars and ruled 
at Istakhr: he was succeeded by his son, Shapur, in 
about 209. Shapur's younger brother, Ardashir I, 
succeeded him in about 216, expanded his rule over 
the rest of Fars, Khuzistan [q.v.] and Kirman [q.v.], 
and, allied with the rulers of Adiabene and Gar- 
makan, defeated and killed the Arsacid Artabanus V 
in battle on the plain of Hormizdagan, probably in 
western Media, in about 224. He commemorated his 
victory in a rock relief near Gur (Firuzabad [q.v.]), 
where he also built the city of Ardashlr-Xwarrah (the 
fortune of Ardashir). In the west, he conquered 
Maysan [q. v. ] , took Ctesiphon, where he was crowned 

in 226, but failed to take Hatra (al-Hadr [q.v.]), con- 
quered Media, and was turned back from Armenia 
(Armlniya [?.».]) by the sons of Artabanus in about 
227. In the east, he occupied the rest of the Parthian 
empire by force and intimidation, extinguishing the 
royal fires of local rulers. He also occupied coastal al- 
Bahrayn [q.v.] and 'Urnan [q.v.] in Arabia. In 230 he 
invaded Roman Mesopotamia, claiming everything 
as far as Ionia and Caria by ancestral right from 
Cyrus to Darius, the last Persian king, whose 
kingdom Alexander had destroyed. Hatra went over 
to the Romans, whose counter-invasion of Armenia 
and Media by the generals of Alexander Severus was 
beaten back by Ardashir with heavy losses on both 
sides. After the death of Alexander Severus in 235, 
Ardashir was able to take Nisibis and Carrhae (Har- 
ran [q.v.]) in 238. In 240 his son, Shapur, was crown- 
ed as co- ruler, possibly to secure the succession. Hatra 
fell to them in 240, and they appear to have been vic- 
torious in Armenia before Ardashir died early in 242. 

Ardashir I was the real founder of the Sasanid 
monarchy. This was initially based on a cult of divine 
kingship that continued from the Arsacids, who had 
been called theos theopator ("god, of divine descent") 
and represented as receiving a diadem from Tyche on 
their coins, and had a royal, dynastic fire dedicated to 
them as to the other gods. On his coins Ardashir's 
earliest crown resembled that of Mithradates II, he 
was called "the Mazda-worshipping god, Ardashir, 
who is descended from the gods", and the reverse 
bore the image of a throne mounted on a column with 
the inscription "the fire of Ardashir". His royal fire 
was kindled at the beginning of his reign in 224-5. In 
rock reliefs, he was shown receiving a ring-like object 
from Ohrmazd, who was also portrayed as a king, 
probably to show the divine nature of royal authority, 
and/or that Ardashir ruled the world on behalf of 
Ohrmazd. Later tradition credits Ardashir with 
having the Avestan teachings collected and adding the 
five Gatha days to the Zoroastrian calendar. 

The Achaemenid legacy was also important for the 
early Sasanid monarchy, beginning with the irreden- 
tist claims of Ardashir (also made by the late Arsacids) 
reported by the Greek and Latin writers (who called 
him Artaxerxes). This legacy may be seen in the 
Sasanid dynastic roots at Istakhr. the use of architec- 
tural motifs from Persepolis in Ardashir's palace at 
Flruzabad, the choice of the major Achaemenid site of 
Naksh-i Rustam for reliefs, inscriptions, and fire 
temples, and the later claim of their descent from 
Dara. Under Shapur I this is seen in his use of an 
Achaemenid building at Naksh-i Rustam for his in- 
scription and in the interest of some of his officials in 
viewing the paintings of Esther and Ahasuerus (Ar- 
taxerxes) in the synagogue at Dura. 

The structure of the Sasanid state under Ardashir 
was outwardly similar to that of the late Arsacids with 
sub-kings for Abrenakh (probably Khurasan [q.v.]), 
Marw [q.v.], Kirman, and the Sakas (Slstan [q.v.]). 
Greater royal centralisation was reflected in his 
establishment of an elaborate royal court attended by 
the sub-kings and the heads of Parthian noble 
families, and his foundation of royal cities in crown 
territories (dast(a)gird) as centres for royal ad- 

Shaptir I continued the work of his father, possibly 
conquering Gilan [q.v.] and Kh w arazm [q.v.] before 
being drawn west by a Roman attack in 243 under 
Gordian III, who took Carrhae and Nisibis, annexed 
Osrhoene with Edessa (Urfa [q.v.]), garrisoned 
Singara (Sindjar [q.v.]), creating the province of 
Mesopotamia, and attacked Babylonia (Asorestan). 

Shapur defeated and killed Gordian at Massice on the 
middle Euphrates, which he renamed Peroz-Shapur 
("victorious is Shapur". also called al-Anbar [q.v.]). 
Philip the Arab made peace with Shapur for 500,000 
denarii and possibly a free hand in Armenia, which 
stabilised the frontier for about ten years. 

In 252 Shapur had the Arsacid king of Armenia, 
Khusraw, assassinated. The latter's son, Tiridates, 
fled to the Romans, while the sons of Tiridates joined 
Shapur. One of them, Artavazdes, ruled Armenia 
from 252 until 262. Claiming Roman harm to 
Armenia, in 256 Shapur attacked the Romans on the 
Euphrates, destroying a Roman army at Barbalissos, 
taking Dura, occupying Mesopotamia, and invading 
Syria, where he took Antioch (Antakiya [q.v.]) with 
the help of a local senator called Cyriades, who took 
the imperial title with Persian backing. Shapur cam- 
paigned up the Orontes valley as far as Emesa (Hims 
[q.v. ]) and returned to Mesopotamia. The Persian 
garrison in Syria was massacred in 257, Cyriades was 
murdered by the people of Antioch, and Dura was 
reoccupied by Roman-Palmyrene forces between 256 
and 260, when Shapur retook it. Shapur settled his 
Syrian captives in Khuzistan at a new city called Weh 
Antiok Shapur ("the Better Antioch of Shapur"), 
which became Djundayshabur (Gondeshapur [q.v.]). 

Between 258 and 260 Shapur besieged Edessa and 
Carrhae and defeated a Roman relief army near 
Edessa, taking Valerian captive. According to Kirdlr, 
who was with the army, Shapur then pillaged, burn- 
ed, and destroyed Antioch and Syria, Tarsus and 
Cilicia, Caesarea and Cappadocia, as far as Greece, 
Armenia, Georgia (al-Kurdj [q.v.]), Albania, and 
Balasagan up to the Gates of the Alans. Except for 
Magians (Madjus [q.v.]), who were not harmed, 
Shapur deported tens of thousands of captives to Fars, 
Parthia, Khuzistan, and Babylonia. He was harrassed 
on his return by Odainath of Palmyra, who retook 
Carrhae and Nisibis, besieged Ctesiphon, recovered 
the booty (although Kirdlr claims the booty was 
returned), captured Shapur's concubines, and 
restored the Roman frontier between 262 and 267. 
Shapur's settlement of his captives was part of the 
economic development of his state through the use of 
captive labour, and a denial of those labour resources 
to the Romans in the provinces from which they 
came. They were used to build dams, bridges, and ir- 
rigation works in Khuzistan, and to build Shapur's 
palace at Blshapur in Fars. They also introduced 
eastern Mediterranean cultural influences where they 

Shapur commemorated his achievements in at least 
seven rock reliefs, and left a major inscription (SKZ) 
on the eastern wall of the Ka c ba of Zoroaster at 
Naksh-i Rustam in Parthian, Middle Persian, and 
Greek sometime after 262, when he replaced Ar- 
tavazdes with his own son, Hormizd-Ardashir, as 
great-king of Armenia (262-72). In his inscriptions 
Shapur was called "the Mazda-worshipping god, 
Shapur, king of kings of Iran and non-Iran, of divine 
descent." There was a royal fire at Blshapur; 
Shapur's coins were the first to have the fire altar 
flanked by two attendants (in the 3rd century some- 
times the king and queen) on the reverse; he founded 
many Bahram fires, of the god of victory, in every 
land; he gave benefices to many Magians; he estab- 
lished name-fires for his own soul and for members of 
his immediate family and daily sacrifices of sheep, 
bread, and wine for twenty-nine members of the royal 
family, living and dead. Several of his sons had the 
names of deities. Such royal patronage of the Magian 
cult, with fires possibly beginning to replace cult 

images, should be balanced by Shapur's reputation 
for tolerance, his patronage of both Kirdlr and ManI 
(who converted some members of the royal family), 
and his toleration of Christian and Jews. The latter 
were left under their exilarch in return for respecting 
Persian law and paying taxes. According to Elisaeus 
Vardapet, in the 5th century, Shapur issued an edict 
that people of every religion should be left undisturb- 
ed in their belief. According to the Denkard (DKM, 
412-13) he collected Magian literature which had been 
dispersed in India, the Byzantine Empire and other 
lands, that dealt with medicine, astronomy, and 
philosophical subjects, added them to the Avesta, had 
a copy put in the royal treasury, and considered 
bringing all systems into line with the Mazdayasnian 
religion. There has been modern speculation that 
Shapur favoured a universalising religious syncretism 
corresponding to his imperial ambition. 

In his inscription (SKZ) Shapur described his em- 
pire as consisting of Fars, Parthava, Khuzistan. 
Maysan, Asorestan (Babylonia), Adiabene, Arabia 
(Beth c Arbaye), Adharbaydjan. Armenia, Iberia, 
Maxelonia (Lazica), Albania, Balasagan up to the 
Caucasus mountains and the Gates of the Alans, the 
Elburz mountains, Media, Gurgan [q.v.], Marw, 
Harat [q.v.], Abarshahr, Kirman, Slstan, Tugran, 
Makran [q.v.], Pardan, Hind and the land of the 
Kushans up to Pashkabur (Peshawar ?), to the 
borders of Kish, Sogdia and the mountain of Shash 
(Tashkent [q.v.]), and Mazon fUman) on the other 
side of the sea. The only definition of the difference 
between Iran and non-Iran is given by Kirdlr, accord- 
ing to whom Iran included Maysan, Adharbaydjan. 
Isfahan [q.v.], Rayy [q.v.], Kirman, Slstan, Gurgan, 
Marw, and as far as Pashkabur in the east. Non-Iran 
was the territory in the northwest including Armenia, 
Iberia, Maxelonia and Balasagan up to the Gates of 

Centralising tendencies under Shapur were ex- 
pressed in three ways. Members of the royal family 
participated in rule from the beginning of the dynasty 
and their appointment as sub-kings began the process 
of converting client kingdoms into provinces. By the 
end of Shapur's reign five of his sons, who did not 
start local dynasties, ruled Slstan, Gilan, Kirman, 
Maysan and Armenia as kings. Second, he continued 
Ardashir's policy by founding at least eleven cities out 
of the expanding royal domain (dast(a)gird), and ap- 
pointed governors (shahrabs), with a staff of scribes, a 
treasurer, and a judge, for at least eight to ten new 
and existing cities. None of these cities were in the ter- 
ritory of sub-kings and thus constituted the core of 
directly administered territory in Asorestan, 
Khuzistan, Fars, Media and Abarshahr (Khurasan). 
Third, administration was centralised at an increas- 
ingly hierarchic and elaborate royal court, attended 
by members of the Varaz, Suren and Karen families, 
who had Sasanid dynastic names but no ad- 
ministrative positions. (Two princes called Sasan were 
also raised by the Farrlgan and Kidugan families.) 
Court officials included the bidaxs (or pitiaxs, viceroy, 
regent?), hazarabed (commander of the royal guard), 
framadar (steward of royal property), chief scribe, chief 
judge, treasurer, and market inspector (wazarbed, 
agoranomos). From the third century the Sasanid ruling 
classes were organized into a hierarchy in descending 
order of kings and queens (shahrddrdn), princes 
(waspuhragan), grandees (wuzurgdn), including power- 
ful families from the Arsacid period and court of- 
ficials, and the minor nobility (azaddn). 

When Hormizd-Ardashir (Hurmuz [q.v.], 272-73) 
succeeded his father, his brother, Narseh, the Sakan- 

shah, became king of Armenia (272-93). Hormizd- 
Ardashir was the first to be called "king of kings of 
Iran and non-Iran" on his coins. He allowed ManI to 
preach, but he promoted Kirdlr to Ohrmazd-mowbed 
(mobadh, [q.v.]), and put him in charge of ritual and 
sealing contracts. The third century was a transition 
period between the Hellenistic tradition of expressing 
political allegiance through the cults of deified rulers 
and the emergence of confessional religions with mass 
memberships that became identified with states. By 
the mid-third century Manichaeism, Christianity, 
and an early form of Mazdaean Zoroastrianism were 
all spreading in Sasanid territory. The success of the 
Mazdaean priests in the late 3rd century was related 
to the career of Kirdlr, the proliferation of Bahrain 
fires, and the need of the family of Bahram I [q.v.] 
(273-76) to legitimise their succession against the 
claims of his brother Narseh. Bahram I had been king 
of Gilan under Shapur; he was shown being invested 
by Ohrmazd in a rock relief at Bishapur; he wore the 
solar crown of Mithra on his coins; and enjoyed 
fighting, hunting, and feasting. When ManI heard of 
his accession, he set out for Kushan territory, but was 
forbidden to go and told to present himself to 
Bahram I at Djundayshabur, where Kirdlr accused 
him to the king of being opposed to hunting and war. 
Mam's answer that he had healed many people and 
driven out demons was fruitless; he was imprisoned at 
Djundayshabur, where he died in 276. His successor, 
Sisinnios, organised the Manichaean ecclesiastical 

Bahram I was succeeded by his son, Bahram II 
(276-93), who was married to Shaparduxtak, the 
daughter of Shapur, king of Maysan. The legitimacy 
of this branch of the royal family was emphasised by 
showing Bahram II with his wife and son, Bahram, on 
the coins, some of which showed the king and queen 
attending the fire altar on the reverse. A close associa- 
tion with the god/yazad Varethraghna (Bahram) was 
shown on the coins by the introduction of the wings of 
Varagn (the eagle), the bird form or symbol of 
Varethraghna and of victory, on Bahrain's crown, 
and by his queen and heir wearing animal forms of 
Varethra gh na and other yazads on their heads, ac- 
cording to Lukonin. Bahram II was depicted in at 
least ten rock reliefs engaged in combat, with his fami- 
ly, holding court, receiving the submission of Arabs, 
killing two lions and attending a fire altar. Bahram II, 
his queen, and his heir were also represented on the 
earliest attested such Sasanid silver cup. 

Kirdir reached the height of his career under 
Bahram II, when he left four inscriptions with similar 
texts, located strategically near rock reliefs of Ar- 
dashlr I, Shapur I and Bahram II. In them he re- 
counted his earlier career and his promotion to mowbed 
and judge of the entire state under Bahram II, who 
also made him master of ritual and put him in charge 
of the fires of Anahita-Ardashir and of Lady Anahita 
at Istakhr. marking the first time the latter position 
was separated from the person of the monarch. In his 
time, the Mazdaean religion (den Mazdesn) and 
Magians were in great honour in the state, and there 
was great service to the yazads, water, fire and cattle. 
He claimed to have destroyed idols and established 
the place of the yazads, increased religious rituals, 
founded many Bahram fires, sealed many contracts 
for the fires and the Magians as well as testaments and 
documents, and arranged many close-kin marriages. 
Many of the unfaithful became faithful and aban- 
doned the doctrines of the demons for that of the 
yazads. He also claimed to have suppressed Jews, Bud- 
dhists, two kinds of Christians, "baptists", and Zan- 

dlks (probably Manichaeans). Manichaeans were 
persecuted under Bahram II, their leader, Sisinnios, 
was killed, but they were protected and patronised by 
c Amr b. c AdI, the amir of the Lakhm (Lakhmids 
[q.v.]) at al-HIra [q.v.]. Kirdlr's claims may have been 
exaggerated, but they mark the first recorded attempt 

affirmation of the Mazdaean identity by opposition to 
other religions in the late 3rd century. 

These developments also contribute to the thesis 
that the Sasanids persecuted non-Mazdaeans when 
the monarchy was weak and needed support from the 
Mazdaean priests. In addition to the Bahrams' 
problem with legitimacy, Bahram II had to fight on 
two fronts. In 283 a Roman army under the emperor 
Carus reached Ctesiphon, but Carus died and the ar- 
my withdrew. However in 287 Bahram II made peace 
with Diocletian in order to deal with the revolt of his 
brother Ohrmazd in the east. According to the terms, 
the Sasanids ceded their claims to upper Mesopotamia 
and western Armenia, and an Arsacid prince was en- 
throned in western Armenia as Tiridates IV , leaving 
Narses with eastern Armenia. By the end of his reign 
Bahram II had defeated Ohrmazd, conquered SIstan 
and made his heir Sakanshah. 

The death of Bahram II in 293 set off a dynastic 
civil war. One noble faction, ledby Wahnam and 
joined by the king of Maysan, Adurfarnbag, sup- 
ported the succession of the Sakanshah as Bahram III. 
Other nobles rebelled, gathered in Asorestan, and 
sent for Narseh, the last son of Shapur, who came 
from Armenia to Iran and met them at Paikuli north 
of Khanikin \q. v.]. There Narseh was proclaimed king 
by an assembly of nobles that included Shapur the tax 
collector (hargubed), Papak the bidaxs, ArdashTr the 
hazarabed, R akhsh the spahbed and the heads of some of 
the great families. This event was commemorated by 
a stone monument bearing a major inscription in 
Middle Persian and Parthian, that includes events 
later than the meeting in 293. In this inscription 
Narseh proclaimed his legitimising propaganda that 
"Shapur willed that this one [Narseh] should be lord 
over Eran-sahr, because he was the best, the most 
just, and the most vigorous after Shapur." He also 
recorded his recognition by twenty-seven rulers, who 
either acknowledged him as king of kings, according 
to Lukonin and others, or acknowledged his right to 
the throne, according to Frye. If the first, it would in- 
second, it would indicate a significant contraction of 
the state since the time of Shapur, and the revival of 
local rulers. Narseh then defeated the king of Maysan 
and rival forces in Khuzistan, captured and executed 
Wahnam (the fate of Bahram III is unknown), and 
replaced the name of Bahram I on the relief at 
Bishapur with his own. 

He was called "the Mazdaean, divine Narseh, king 
of kings of Iran and non-Iran, descended from the 
Gods" on his coins, which had his own face in the 
flames of the fire-altar on the reverse. In a rock relief 
at Naksh-i Rustam he was depicted receiving a ring 
from a female figure, usually identified as Anahita, 
and at Paikuli he claimed to have recovered Armenia 
for the land of Iran through Ohrmazd, all the yazads 
and the Lady Anahita. Beneath the common venera- 
tion of Ohrmazd one can see a shift from one yazad to 
another, from Verethraghna to Anahita, associated 
with the new branch of the dynasty. Narseh reversed 
the policies of the Bahrams, neglected fire temples, 
perhaps in favour of his own royal fire (his eldest 
grandson was called Adur-Narseh "the fire of 
Narseh"), and tolerated Manichaeans. 

In 296 Narseh conquered Armenia, invaded 
Mesopotamia, and prepared to invade Syria. Diocle- 
tian sent the Caesar Galerius, whose forces were 
crushed by Narseh. Galerius regrouped at Antioch 
and invaded Armenia in 297 where he routed Narseh 
and captured his family, while Diocletian advanced to 
Nisibis. In 298 Narseh secured the return of his family 
by ceding several districts on the upper Tigris to the 
Romans and setting the frontier at the Tigris. 
Tiridates was restored in Armenia, which was enlarg- 
ed in the east to the frontier of Media. The king of 
Iberia was to be invested by the Romans, and Nisibis 
was to be a place of commercial exchange. The treaty 
of Nisibis thus marked a change in the balance of 
power at the end of the 3rd century with the recovery 
of the late Roman empire under Diocletian. 

Narseh was succeeded by his son Hormizd II (302- 
9), who continued his father's policies and was re- 
presented in combat in a rock relief at Naksh-i 
Rustam. It was during his reign that Armenia became 
Christian under Tiridates IV, making Christianity a 
political issue for the Sasanids, especially after the 
conversion of Constantine in 311, and dividing 
Armenia henceforth between pro-Roman Christians 
and pro-Sasanid nobles. 

Hormizd II had several sons; the eldest, Adur- 
Narseh took the throne briefly when his father died in 
309, but the officials and priests at court made a 
posthumous son, Shapur II (309-79), king and en- 
throned him as an infant. Shapur II's minority lasted 
for fifteen years; his reign lasted for seventy years, the 
longest of any Sasanid king. The survival of the 
Sasanid state during Shapur II's minority indicates 
the strength of ruling institutions by the 4th century 
and the fact that the monarchy had become more im- 
portant than the monarch as a focus of loyalty for 
nobles, officials and priests who had a vested interest 
in the system. The state was held together by its ad- 
ministrative structure, and a glimpse of local officials 

scriptions on a doorpost at Persepolis from 311 and 
327. According to post-Sasanid accounts, the main 
problem during Shapur II's minority was the incur- 
sion of Arabs from Eastern Arabia into c Irak [q.v.], 
Khuzistan and the coast of Fars. At his majority in 
324 Shapur II drove them out of c Irak, began to build 
long walls and a trench (khandak Sdbur) along the desert 
frontier in the south-west to keep them out, sailed 
from Fars to al-Bahrayn, defeated the Arabs there, 
made eastern Arabia a province with garrisons and of- 
ficials and may have built the Sasanid fort at Siraf 
[q.v.] as a naval base. He resettled Arabs in 
Khuzistan, Fars and Kirman. 

The resumption of war with the Romans over 
Armenia and Mesopotamia from 337-8 to 350 coin- 
cided with the first persecution of Christians in the 
Sasanid state and support for Shapur by anti- 
Christian, pro-Sasanid Armenian nobles. In about 
350 Shapur II had the Armenian king Tiran killed 
and may have made the latter's son, Arshak (ca. 350- 
67), king. At about the same time, Christians rebelled 
at Susa; Shapur II had the city razed and trampled by 
elephants and built the new city of Eran-Xwarrah 
Shapur ("Iran's glory [built by] Shapur") nearby. 
The persecution lasted until the end of his reign, en- 
couraging the thesis that henceforth Christians tended 
to be persecuted during wars with Rome. 

At mid-century Shapur was drawn east by the inva- 
sion of Central Asia by Chionite Huns. Iranians mix- 
ed with Huns were driven south and attacked eastern 
Iran. In the 350s Shapur II campaigned in the east 
against the Chionites and Kushans. The Kushans 

were conquered and their territory governed into the 
early 5th century by a series of nine Sasanid royal 
princes with the title of Kushanshah according to 
Lukonin and Gobi, although there are good reasons to 
put this in the 3rd century. Shapur II came to terms 
with the Chionites, concluding a treaty of peace and 
alliance with them and the people of GTlan (Gelani) in 
about 358. 

He then returned to the Roman front with 
Chionite, Gelani, Albani and Segestani allies. In a let- 
ter to Constantius he revived Achaemenid irredentist 
claims to territory as far as the river Strymon and the 
border of Macedonia, as well as saying that it was his 
duty to recover Armenia and Mesopotamia, which 
had been taken from his grandfather, Narses, by 
double-dealing. He took Amid in 359 and Singara in 
360, deporting the population to Persia. In 363 Julian 
marched down the Euphrates and invaded Babylonia, 
while Armenian forces marched east from Carrhae. 
Julian's rebuilding of the temple in Jerusalem encour- 
aged Jews at Mahoza (Weh-Ardashlr, across the 
Tigris from Ctesiphon) to believe the Messiah had 
come to return them to Palestine; thousands of them 
marched out of the city and were massacred by 
Shapur's troops. Ammianus Marcellinus, who was 
with Julian's army has left a first-hand account of his 
campaign and a description of Babylonia. He des- 
cribed wall-paintings of hunting scenes in a palace 
near Ctesiphon, an annual fair at Batnae in Fars at 
the beginning of September to buy goods brought by 
Indians and Syrians by land and sea, and cities and 
villages all along the coast of the Gulf where many 
ships sailed back and forth. He also described the 
battle-order of the Sasanid army at Ctesiphon with 
mail-clad cavalry in front, infantry behind them and 
elephants in the rear. The Romans won the battle at 
Ctesiphon in 363, but Julian was killed, and Jovian 
secured the retreat by ceding Nisibis, Singara and the 
trans-Tigris districts that Narseh had lost, and aban- 
doning Armenia. Between 368 and 370 Shapur took 
Armenia by a combination of diplomacy and force, 
won over some nobles and satraps, invited the king, 
Arshak, to a banquet, blinded, imprisoned, and killed 
him, pillaged Armenia, drove the Roman protege out 
of Iberia and enthroned his own one. Roman counter- 
measures left Iberia divided and Armenia under a son 
of Arshak as a Roman protege, whom they had killed 
in about 373. 

The reign of Shapur II marked the beginning of a 
shift in royal legitimising ideology from being of 
divine nature and descent to being human and 
descended from the Kayanid [q. v. ] kings mentioned in 
the Avesta. Shapur II was the last Sasanid king to be 
called "of divine descent" on his coins. In his letter 
to Constantius, in 358, he called himself "king of 
kings, partner with the stars, brother of the sun and 
moon." The transition was symbolised by emphasis 
on the royal fortune, Hvarnah (xwarrah, fan), as a 
yazad the subject of a Yasht, represented by a ram with 
a bow around its neck. At the siege of Amid, in 359, 
Shapur wore a gold ram's head set with gems. The 
garrison of Mahoza, in 363, extolled in song the 
justice and good fortune (iustiliamfelicitatemque) of their 
king. He ordered the Christian, Post, to venerate the 
"fortune (gad) of Shapur, king of kings, whose nature 
is from the gods." According to Lukonin, the dynastic 
myth of Kayanid descent through Dara was invented 
in Shapur's time by a secretary, Khorkhbud. who 
wrote a chronicle telling how the Hvarnah of the 
Kayans had watched over Ardashlr I in the form of a 
ram with a bow around its neck, ensuring his victory 
over the Parthians. 

The 4th century also saw a shift from rock reliefs to 
royal silver objects with hunting scenes. The metal 
came from a single source and the objects were proba- 
bly produced in a royal workshop beginning with 
ghapur II, such as that run by Post the karugbed at 
Eranshahr-Shapur (Karkhe dh a Ladhan). Shapur II 
was also associated with royal banqueting customs 

This same century was a period of sharpening 
religious conflict, inter-faith polemic and the con- 
solidation of religious identities. The polemic lines 
were drawn among Magians, Manichaeans, Christ- 
ians, Jews and fatalists. Magians and Christians op- 
posed Manichaeans and fatalists; Magians and Jews 
opposed Christians, probably indicating who felt 
threatened by whom. According to the Denkard (iv, 
26-7) Shapur II (who may have been Zurvanite) 
issued an edict defining true belief and condemning 
false belief, but does not say what they were. Under 
Shapur II the rod, Mahraspand, sanctioned the con- 
fiscation of the property of Manichaeans for the royal 
treasury. The latter's son Adurbed was a major figure 
in the formulation of Mazdaean doctrine, remem- 
bered and quoted in later tradition. He stood for 
judicial procedure, an ethical reduction of the sphere 
of life controlled by fate and the embodiment of the 
spiritual in the material, and he opposed 
ml ead of popular piety was 

reflected in the motto "i 
sonal seals, although the me 
ning to shift from the divi 
term for Ohrmazd. 

The organisational 
ians and Jews underwent elaboration in this centur 
The earliest references to a supreme mowbed and to 
priestly hierarchy with territorial jurisdictions corr 
sponding to secular administrative divisions belong 
ign of Shapur II. As part of the ruli 

the yazdan 
;ng oi yazdan was begin- 
beings in general to a 

of Magians, Christ- 


s and the 


of fire 

■ing the poll tax. Chi 
formed an ecclesiastical structure under the primacy 
of the bishop of Ctesiphon from the early 4th century, 
which made it easier for the state to deal with them. 
Shapur II required the bishop of Ctesiphon, Shem'on 
bar Sabba c e, to collect a double poll tax and land tax 
from Christians because they were living in peace. 
Jews were ruled indirectly through their exilarch in 
return for taxes, loyalty and defence. Subordinate of- 
ficials of the exilarch administered justice, collected 
taxes for the government, supervised markets, pre- 
served order and defended Jewish towns. The 4th- 
century exilarch, c Ukba bar Nehemiah, established 
the principle that the law of the government was to be 
obeyed just as the Torah. By the end of the 4th cen- 
tury, there was an alliance between the exilarch and 
the rabbis, who staffed the religious courts, where 
they applied religious law to the life of the community, 
acted as local administrators, collected the poll tax and 
enforced dietary laws as market inspectors. By this 
century world-affirming doctrines were being used by 
the representatives of authority in the new order: 
Mazdaean priests, Jewish rabbis and the Christian 
clergy, who used ritual and law to exercise social 

s of S 

r II a thre. 

Under the 

power struggle emerged among the monarchy, the 
wuzurgan or nobles, and the Magian priests, while the 
pressure from steppe peoples increased. The Sasanids 
were generally successful in warfare in the east until 
the mid-5th century. Ardashir II (379-83) was most 
probably the brother of Shapur II; an investiture relief 
at Tak-i Bustan near Kirmanshah [q.v.] is usually 

ascribed to him, although it could be that of Shapur 
II. He is said to have been killed in a revolt by the 
wuzurgan, and was succeeded by Shapur III (383-8), a 
son of Shapur II, who is said to have favoured the 
wuzurgan. Shapur III was the first to be called 
Kayanid on his coins, and was represented as slaying 
a lion on a silver plate; he left a rock relief of himself 
and his father at Tak-i Bustan. He is said to have 
freed Christian prisoners because it was more pro- 
fitable to the state to have them engage in crafts and 
pay taxes, which may have led the wuzurgan to replace 
him with his brother Bahram IV (388-99). Some of 
the latter's coins have the earliest mint-marks. In 389 
Armenia was partitioned with the Romans, at first 
under two Arsacid client kings, but in the 390s the 
western part was integrated into the Roman empire 
and governed by a comes, while Bahram IV made his 
brother Bahram-Shapur/Vramshapuh king of Perse- 
Armenia in 394, marking the end of the Arsacid 
dynasty of Armenia. In 395 Huns from north of the 
Caucasus reached Mesopotamia and were driven 
back. Bahram IV died violently, possibly at the hands 
of the wuzurgan. Frye argues that the strength of local 
nobles over the central government in the late 4th cen- 
tury was due to Shapur II's failure to establish central 
authority, although it is equally possible that Shapur's 
wars served to strengthen the landed military nobility, 
officials, and royal princes. 

Yazdadjird I (399-420) was most likely a brother of 
Bahram IV and was called "delight of the state" on 
his coins. The incipient outline of a dual secular and 
religious judicial system existed by the tenth year of 
his reign (408-9). There was a secular hierarchy of 
state judge, local rod and judges of first instance, while 
the priests, under the gjand Magus, performed 
judicial and ritual duties. Adur-Ohrmazd (d. 407) is 
said to have recited the A vesta, Yashts and drdn 
(ceremonial food offerings) night and day as mowbed of 
Belashfarr. The expansion of the royal do] 

archie admi 

n the. 

attestation of a grand-framdddr in 410, who was suc- 
ceeded by the famous Mihr-Narseh, from the noble 
clan of Spendiad, who dominated the first half of the 
5th century. The latter left an inscription on a bridge 
he had built at Firuzabad, and, according to al- 
Tabari, developed four villages with orchards nearby, 
founded a sacred fire for his own soul in one and 
"named fires" in the other three for each of his sons: 
Zurwandad, the Herbed of herbeds, Kardir, the head 
of soldiers and Mah-Gushnasp, the head of 
cultivators/herdsmen. These titles were new in the 5th 
century; two of them are Avestan; and all three 
categories are Avestan. Rather than indicating the 
reorganisation of late Sasanid society according to 
Avestan social categories, they probably show the im- 
pact of Avestan social theory in the 5th century co- 
ordinated with the late Sasanid system of three major 
fire temples: Adur Farnbag, of priests in Fars; Adur 
Gushnasp, of soldiers at ShTz [q.u.] (Takht-i 
Sulayman) in Adharbaydjan: and Adur Burzen Mihr, 
of farmers in Khurasan, that probably existed by the 
5th century. 

The late 4th and early 5th centuries marked the 
height of the alliance between the Sasanid state and 
Jewish exilarchs, who resembled client kings. The 
wife of Yazdadjird I is said to have been the daughter 
of an exilarch (Yahuddn-Shah); at her request he settled 
Jews at Gay (Isfahan [q.v.]). He is said to have 
honoured the exilarch, and Rabbi Ashi regarded the 

Yazdadjird I also legitimised the position of Christ- 
ians. After peace with the Romans/Byzantines in 409, 

Christians were allowed to worship publicly and bury 
their dead. In 410 the first synod in Sasanid territory 
was arranged by the gr and- framaddr at Seleucia/Weh- 
Ardashlr, ' where the creed of Nicaea (325) was ac- 
cepted and a hierarchy of metropolitan bishoprics 
under the catholicos of Seleucia-Ctesiphon was estab- 
lished parallel to the secular administration. The acts 
of this council were guaranteed by the double sanction 
of excommunication and of punishment by the King 
of Kings, and Yazdadjird was called "victorious" and 
"illustrious". Mazdaeans called him a sinner (athim 
in the Arabic sources) and remembered him as an op- 
pressive, cruel tyrant, which probably reflects the op- 
position of priests to toleration and of the wuzurgan to 
royal authority. Toleration was tested toward the end 
of his reign when a Christian priest in Khuzistan 
destroyed a fire temple and refused to rebuild it, and 
another extinguished a sacred fire and performed the 
eucharist in its place. Both were executed after trials, 
but rather than amounting to a "persecution", this 
served to define the limits of toleration. According to 
legend, the wuzurgan prayed for relief from Yaz- 
dadjird's oppression, and God sent an angel in the 
form of a horse that killed him. 

The wuzurgan tried to exclude the sons of Yaz- 
dadjird from the succession. His eldest son, Shapur. 
who had succeeded Vramshapuh in Armenia in 414, 
went to Ctesiphon and took the throne, but was killed 
by the wuzurgan, who enthroned a descendent of Ar- 
dashir I called Khusraw. Shapur's brother Bahram, 
who had been raised at al-Hira by al-Mundhir of 
Lakhm, marched on Ctesiphon with an Arab army; 
Khusraw abdicated and Bahram seized the throne as 
Bahram V (420-38). He remitted taxes at festivals, 

his reign he defeated an invasion of Hephthalites 
(Hayatila [q.v.]) in the east, killed their king, 
dedicated his crown to the Gushnasp fire at Shlz and 
the queen and her slaves to serve there. In the west, 
war was waged with the Byzantines from 421 to 422 
by Mihr-Narseh over the extradition of Armenian 
Christian refugees, while, at the instigation of Mihr- 
Shapur, head of the Magians, for five years corpses 
buried in the time of Yazdadjird I were exhumed and 
scattered about in the sun. The treaty with the Byzan- 
tines in 422 provided for religious freedom to Christ- 
ians in the Sasanid state and to Magians in the Byzan- 
tine state, and for the Byzantines to help pay for the 
defence of the pass at Darband [see bab al-abwab] by 
the Sasanids. A synod in 424 established the 
autonomy of the Christian Church in Sasanid ter- 
ritory from the Byzantine Church. 

Artashes, the son of Vramshapuh, had succeeded 
Shapur as king of Armenia, but the pro-Persian 
naxarars asked to have him removed, so Bahram V re- 
placed him with a Persian marzban (marzpan [q.v.]), in 
428. Following an experiment in the 4th century, the 
marzban emerged during the 5th century as a new kind 
of military governor for a frontier region, replacing 
royal princes as sub-kings. By the end of the 5th cen- 
tury, Nisibis and Asorestan were governed by 

Against Christians, who accused Magians of being 
polytheists and worshipping the elements, Bahram V 
answered that he recognised only one deity, the rest 
were only courtiers of the king. He devoted Mihr- 
Narseh "as a slave" to the Ardwahisht fire and that 
of Afson-Ardashir for several years, and tradition 
associated him with the Adur Gushnasp shrine at 
Shlz. When he married an Indian princess, who 
brought the port of Daybul [q.v J as her dowry, he en- 
trusted her to the priest of the Adur Gushnasp shrine 

to be purified and converted to Magianism, and he is 
said to have visited this shrine on the feasts of Sada 
(the midwinter fire festival) and Nawruz [q.v.]. 

Tradition also associated Bahram V with importing 
Indians (Zutt [q.v.]) with water buffaloes to the Gulf 
coast of Fars, and 4,000 Indian 
spread through the provinces ; 
remembered as Bahram Gur ("the onager") for his 
skill in hunting, and his exploits were extolled in 
legend and illustrated in art. 

For a half-century after Bahram V there was grow- 
ing disorder in the Sasanid state, persecution and 
forced conversion of Christian and Jews, a seven-year 
drought and famine in the 460s or 470s, the payment 
of costly tribute to the Hephthalites and social 
disorders associated with the Mazdakite movement 
[see mazdak] at the end of the century. The shift from 
divine to Kayanid kingship was completed by the time 
of the son of Bahram V, Yazdadjird II (438-57), the 
last to be called "divine" on his coins, while the title 
' ' Kay ' ' was common on his coins and those of his suc- 
cessors. From the time of FTruz (459-84), Kayanid 
names became popular in the royal family. The 
significance of this lay in the role of Kayanid kings as 
patrons of Zoroaster and supporters of his religion in 
the Avesta and in the development of the myth of the 
Kayanid descent of the Sasanid kings included in the 
Book of Kings (Xwaddy ndmag) that contained material 
from at least the 5th century. 

War with the Byzantines was ended peacefully in 
442 in order to deal with the steppe people in the 
north-east. There were repeated campaigns to keep 
the Kidarite Huns out of Gurgan, and Yazdadjird II 
is said to have moved his court to north-east Iran for 
seven years to hold back the Hephthalites. His reign 
also saw the beginning of serious efforts to spread 
some form of Magianism and establish a uniform 
religious identity among Sasanid subjects similar to 
contemporary efforts by the Byzantines to deal with 
the Nestorian and Monophysite controversies. The 
title of mowbedan mowbed (a caique on Shahan-shah) was 
first attested for Mihr-Shapur in 446. Yazdadjird II is 
said to have considered it a sin to accept tribute from 
Christians; in 446 Christians denounced by Mani- 
chaeans were executed at Kirkuk [q.v.]; and the king 
of Albania was forced to become a Magian. He tried 
to convert Armenian Christians and Jews and sent 
Mihr-Narseh to do it, who presented Magian doctrine 
to them in its Zurvanite form. These efforts provoked 
a revolt by Christian Armenians led by Vardan, a 
Mamikonian prince, which was crushed by the pro- 
Sasanid naxarars under Vasak, prince of Siunik, at the 
battle of Avarair in 451 . Afterwards Mihr-Narseh was 
removed from service to the fire temples for sinfulness 
and assigned to the royal domains for several years. In 
455 Yazdadjird II is said to have outlawed the Jewish 

The death of Yazdadjird II was followed by a 
dynastic civil war between two of his sons, Hormizd 
III (457-9) and FTruz (459-84), who defeated and 
killed his brother. During the civil war the Albanian 
king rebelled and allied with the Huns north of the 
Caucasus. After war with the Albanians, FTruz made 
peace, permitting Albanians and Armenians to re- 
main Christian, and the Byzantines subsidised the 
defence of the Caucasus passes. Upon consulting with 
Martbut, the mowbedan mowbed, and other judges, 
FTruz put Mihr-Narseh in the service of the Ohrmazd- 
FTruz fire (i.e. his own royal fire). In- the 460s the 
Hephthalites crossed the Oxus, and in about 469 
defeated and captured Firiiz, who lost Harat to them, 
agreed to pay them tribute, left his son Kubadh as a 

hostage to guarantee its payment and levied a general 
poll tax over his entire state to ransom him. The great 
drought and famine led him to remit taxes, but 
together with the beginning of persecution under Yaz- 
dadjird II, may have encouraged messianic expecta- 
tions among Jews in 468, four hundred years after the 
destruction of the Temple. About that time, the Jews 
of Isfahan are said to have flayed two kerbeds, and 
FTruz ordered half the Jews of Isfahan killed and their 
children to be turned over to the Sros Aduran fire 
temple as slaves. In 470 two rabbis and the son of the 
exilarch were imprisoned and executed, and in 474 or 
477 Babylonian synagogues were closed, schools 
destroyed, Jews were made subject to Persian law and 
Jewish children were turned over to Magians to be 

When FTruz died fighting the Hephthalites in 484, 
the wuzurgdn enthroned his brother Balash (484-8), 
who secured peace with the Hephthalites in return for 
tribute, gave Armenian Christians religious freedom 
and made Armenia a royal province. By the late 5th 
century, Sasanid coins began to have dates in the 
regnal year of the king. The desire of Balash to build 
baths in the cities of his empire is said to have provok- 
ed opposition from the Magian priests, who accused 
him of trying to abolish their laws. In 488 he was 
deposed by the wuzurgdn and priests, who enthroned 
Kubadh I (488-96, 499-531), the son of FTruz. 

Sectarian fragmentation and conflict among 
Magians and Christians were typical of the end of 
Late Antiquity. The 5th century saw the emergence of 
the egalitarian movement of the Zaradushtagan, that 
captured the Sasanid state at the end of the century 
and survived as a sectarian form of Magianism. Con- 
temporary with the Monophysite movement in the 
Byzantine empire, this was related to political support 
for some particular form of the dominant faith, the en- 
forcement of religious conformity for political reasons, 
the popularisation of Magianism through forced con- 
version and the impact of the Avesta, which sanction- 
ed the sharing of wealth, women, and wisdom. 
Drought, famine and the decimation of the wuzurgdn 
in the Hephthalite and civil wars may have created an 
agricultural and demographic crisis in the state, while 
the wuzurgdn engaged in local oppression under a 
weak monarchy. Kubadh I allied with the Zaradush- 
tagan and their leader Mazdak [q.v.] as a means of 
popular support against the wuzurgdn and the priests. 
He tried to force the Armenians to convert, permitted 
or ordered the sharing of women and redistributed 
land. The sharing of women and property may have 
been intended to undermine the wuzurgdn, break 
down social barriers to marriage between nobles and 
commoners, repopulate the state and restore 
agriculture, but people were allowed to help them- 
selves, causing disorders in 494-5. In 496 the wuzurgdn 
deposed Kubadh over his policy toward women and 
held an assembly to decide what to do with him. Pro- 
copius reports that it was not legal to enthrone a com- 
moner unless the royal family was extinct and that the 
majority were unwilling to kill a member of the royal 
family, so they imprisoned Kubadh and enthroned his 
brother Djamasp (496-99). 

Kubadh escaped to the Hephthalites, whose army 
restored him to the throne in 499. Djamasp was blind- 
ed and imprisoned because, according to Procopius, 
no-one with a physical deformity could be king. 
Kubadh's second reign began with a series of natural 
disasters: famine and flooding on the lower Tigris, 
which changed its channels below Kaskar, creating 
swamps and turning land into desert along its former 
course. Needing cash to repay the Hephthalites, the 

Byzantine refusal to pay for the defence of the 
Caucasus passes served as a pretext for predatory war- 
fare against them. In 503 Kubadh took Amid and car- 
ried off the survivors as slaves. The Byzantines made 
peace in 506 and paid an indemnity to Kubadh, who 
returned the conquered territory and the captives 
from Amid. The internal situation was stabilised with 
the Mazdakites in control for thirty years. After his 
restoration, Kubadh required the Arab ruler of al- 
HTra al-Mundhir III (ca. 505-54) to become a Maz- 
dakite. When the latter refused, Kubadh got the Arab 
ruler of the Kinda [q.v.], al-Harith b. c Amr, whose 
kingdom was at its height in the early 6th century, to 
impose Mazdakism on the Arabs in Nadjd and the 
Hidjaz [q.vv.]. In 525 al-Harith defeated al-Mundhir 
and occupied al-HIra until his death in 528. 

Apart from the suppression of the wuzurgdn, who 
never recovered their power in the state, the Maz- 
dakite period was less of a break with than a continua- 
tion of 5th-century developments. The interest in con- 
version and spreading Magian observance in this 
period also saw the emergence of herbeds. as Magian 

couraged to return to their villages to instruct people 
and establish schools, where they taught children to 
recite the Yashts. Mazdak is said to have persuaded 
Kubadh to extinguish all but the three original sacred 

overland and I 

which w 

; Mazdakites may have contributed to theii 
elevation in the late Sasanid period. Kubadh's reputa 
tion does not seem to have suffered, and he wa: 
credited with a reform of the Magian calandar 
Kubadh gave his sons Kayanid names and was callec 
"Kay" on his coins along with increasing astra 

Kubadh also began the reforms associated with the 
recovery of the Sasanid state in the 6th century. H( 
instituted the use of an official seal for mowbeds to us< 

from the late Sasanid period. He began the survey o 
agricultural land in lower c Irak for tax purposes, ant 
began the creation or reorganisation of administrate 
districts later called in Arabic kuwar, and subdistrict: 
called tasog, which became the basis of late Sasank 

The s 
Church in Sasa: 

: pent 

e Chris 

with the introdu 
"western" controversies. This is usually put in terms 
of the formation of a Nestorian majority and 
Monophysite minority, but actually involved the 
adoption of the diophysite christology of Theodore of 
Mopsuestia, favoured by Bar Sawma, bishop of 
Nisibis, who convinced Flruz that, if the Sasanid 
Church adopted Theodore's doctrine, there would be 
less reason to fear Christian collusion with the Byzan- 
tines. He was excommunicated in 485 by his own 
catholicos, Acacius (484-96), who favoured the creed 
of Chalcedon (451). The doctrine of Theodore was in- 
stitutionalised at the school of Nisibis, founded under 
Bar Sawma's patronage in 496, and that year Musa, 
the Christian court astrologer of Djamasp. got royal 
permission (for the first time) for the election of his 
relative Babai as catholicos (497-515), who removed 
Bar Sawma's excommunication at a synod in 499. By 
the 6th century, Nestorians controlled the Church and 
spread Theodore's doctrine while those opposed to 
them tended to become Monophysite. The Armenians 
rejected the Council of Chalcedon in 491, adopted a 
christology close to that of the Monophysites and 
broke completely with the Byzantine Church by 609. 
The late Sasanid state emerged in the 6th century 
with a centralised bureaucratic administration, 

:ial order, commercial domination of the 
an trade, and agricultural 
ent. The revival of a strong 
monarchy, military power, and growing universal 
claims and ambitions culminated in the wars of 
Khusraw II in the early 7th century. This was aided 
by the decline of the Hephthalites, who split into 
eastern and western halves in about 515, and were 
defeated in India in about 528; they survived as small 
states in the east. Kubadh secured the trans- 
Caucasus, drove the pro-Byzantine king out of 
Georgia and installed a marzbtin at Mtskheta in 523. 
War was renewed with the Byzantines in 527 with 
Arab forces under al-Mundhir III, who had recovered 
al-HIra for the Lakhmids when al-Harith b. 'Arar 
died in 528, joining the Sasanid army. In 531 joint 
Sasanid-Arab forces defeated the Byzantines at 
Callinicus, but their own losses were so great that they 

The Mazdakites favoured the succession of Kawus, 
so his younger brother Khusraw, allied with the Maz- 
daean priests, challenged Mazdak's influence over his 
father, assembled the Mazdakites at Ctesiphon for a 
religious disputation or to proclaim Kawus as suc- 
cessor in 528 or 529, convinced Kubadh that Maz- 
dak's doctrines were false and had the latter executed 
with 80,000 followers. When Kubadh died in 531, 
Kawus claimed to succeed his father but, according to 

assume royal power. They were assembled, a docu- 
ment was read declaring Khusraw to be Kubadh's 
successor and they declared Khusraw king. This is the 
first attested example of such a procedure, which was 
the basis for theoretical generalisations in the Testa- 
ment of Ardashir and the Letter of Tansar, both 6th- 

Khusraw I (531-79) purged his rivals among the 
royal family and the wuzurgdn, suppressed the Maz- 
dakites, and ruled fairly unopposed. There are recurr- 
ing references to assemblies during his reign, where 
policies were approved. In 532 he made peace with 
the Byzantines by evacuating fortresses in Lazica, 
while the Byzantines resumed their subsidy to defend 
the Caucasus passes. Khusraw I fortified the pass at 
Darband and built a long wall to defend the Gurgan 
plain. He was famous for his justice and support of 
Mazdaean religion, was remembered as Anushirwan 
[?.».] ("of immortal soul") and became the subject of 
legend. His coins were the first to have stars between 
the horns of crescents, and from 535 had the slogan 
"may he prosper" (abzon). The capital at Ctesiphon- 
Weh-Ardashir grew into a metropolis (al-Mada>in 
\q. v. ]) of several cities, and it was most likely Khusraw 
I who built the great audience hall (Iwan or Tak-i 
Kisra) there. 

The distinction between nobles and commoners 
was restored, the property of Mazdakite leaders was 
confiscated and given to the poor, and property taken 
by force was returned to its former owners. Wuzurgdn 
kept their status but lost their power to a bureaucratic 
elite of royal officials who collected taxes directly. A 
new class of military landlords (dahigdn, dihkdn [q.v.]) 
was created from the azaddn, who were given land by 
the treasury in return for military service and forwar- 
ding taxes, as a basis for the army and support for the 

The state was divided into four quarters, of the 
South, West, North and East, each under a military 
governor (spdhbed), and divided into districts and sub- 
districts, with marzbdns in frontier districts. An 
elaborate central administration run by officials (kdr- 
ddrdn) was divided into departments with parallel 

hierarchies reaching down into the provinces along 
vertical lines of authority with overlapping checks at 
the local level and books on the duties of officials and 
mowbeds. Confidential officers were used as internal 
spies and messengers, and there was a royal post for 
communication. The heads of departments were 
members of the royal court along with the grand 
chamberlain, grand counsellor, head steward, head of 
servants, royal warden, royal astrologer, head physi- 
cian, chief of craftsmen and chief of cultivators; these 
titles were sometimes combined and may have been 
honorific. There were at least seven departments: the 
chancellery, the registry department to seal 
documents, the finance department to collect taxes, 
the department to administer the royal domain, the 
judiciary, the priesthood, including the office of 
religious works (divan-i kardagdn) that registered the 
endowments and property of fire temples and the ar- 
my. A tax reform paid for the expanded system. 
Khusraw I finished the cadastral survey in the Sawad 
[q. v. } of c Irak and replaced the agricultural tax as a 
proportion of the harvest by a tax per unit of area 
under cultivation according to the type of crop, com- 
puted in cash, but paid in cash or kind. He also estab- 
lished a regular, annual poll tax of 4, 6, 8, or 12 
dirhams, to be paid in three instalments, on the male 
population between the ages of 20 and 50. Members 
of the royal family, wuzurgdn, soldiers, priests and 
royal officials were all exempt and agreed to its im- 

A highly developed legal literature was produced in 
the late Sasanid period covering legal procedures and 
social, economic and criminal activity. The Dadestdn 
ndmag was a collection of judicial decisions; the 
Madagan was a digest containing the opinions of jurists 
and royal rescripts from the 5th century until the 
twenty-sixth year of Khusraw II (616). Based on 
Magian ethical principles and religious requirements, 
the application of this legal system helped spread a 
Magian way of life. 

The late Sasanid social ethic emphasised order, 
stability, legality and harmony among the theoretical 
four estates of priests, soldiers, officials and workers, 
so that each would perform its specific duty toward the 
others. These estates were supposed to be hereditary, 
but in practice were overlapped by a vertical social 
hierarchy of the royal family, wuzurgdn, provincial 
governors, small military landholders and local of- 
ficials, freemen and slaves. There are scattered 
references to organisations of artisans and merchants. 
The Sasanids monopolised the transit trade, exchang- 
ing goods with the Byzantines at markets along the 
western frontier at Nisibis and Dubios in Pers- 
armenia. Procopius describes how numerous mer- 
chants from all over Persia and some under Byzantine 
rule came to Dubios and traded for goods from India 
and Iberia. 

Khusraw I was famous for his love of literature and 
philosophy; works were translated from Greek, Syriac 
and Sanskrit into Middle Persian, a set of royal 
astronomical tables (Zidjri Shahi) was produced for 
him and there were philosophical discussions at his 
court. Paul the Persian, a Nestorian theologian and 
Aristotelian philosopher, in a Syriac introduction to 
logic addressed to Khusraw, argued that knowledge 
was better than belief based on the relativity of 
religious belief that resulted in a variety of opinions. 

War was resumed with the Byzantines from 540 to 
561 over the control of Armenia and Lazica, encour- 
aged by Ostrogothic ambassadors, for plunder, and to 
divert and employ the military aristocracy. In 54-0 
Khusraw I invaded Syria, took Antioch, burned the 

huge ai 

leported the survivors, and seized or extorted 
aunts of gold and silver there and from other 
cities on his return. He resettled the captives from An- 
tioch in the new city of Weh Antiok Khusraw in the 
south-eastern part of al-Mada'in, which was pattern- 
ed after Antioch in Syria with public baths and a hip- 
podrome and put under the charge of a chief of ar- 
tisans (karugbad) as a manufacturing centre. In 541 the 
Lazes offered Khusraw I the prospect of attacking the 
Byantines by sea through their land, and he invaded 
Lazica with Huns as allies and established direct rule 
there. In 542 he invaded Syria up the west bank of the 
Euphrates as far as Commagene. Finding nothing left 
to plunder or extort, he deported farmers from 
Callinicus on his return. He defeated the Byzantines 
in Armenia in 543 and invaded Mesopotamia in 544; 
the Byzantines agreed to a five-year truce and paid 
2,000 pounds of gold. The truce was broken in its 
fourth year when the Lazes allied with the Byzantines 
to expel the Sasanids. The fall of Petra to the Byzan- 
tines in 551 was followed by a second five-year truce, 
and negotiations for a peace treaty were begun in 556. 
Khusraw turned to the east, where the Western Turks 
invaded Central Asia and defeated the Hephthalites. 
In about 557-8 Khusraw took the Hephthalite ter- 
ritory south of the Oxus. A fifty-year peace was con- 
cluded with the Byzantines in 561, by which the 
Sasanids evacuated Lazica in return for an annual 
payment of gold, religious freedom was guaranteed 
for Christians under the Sasanids and Magians under 
the Byzantines, and Nisibis and Dara were confirmed 
as centres for the silk trade. Khusraw's gold coins of 
564 contained universalist slogans: "he who makes 
the world without fear" and "may he cause the world 
to prosper". In 564 a member of the Suren family was 
made governor of Armenia, who built a fire temple at 
Dvin (Dwin [q. v. ]) and killed a leader of the Mamiko- 
nian family, provoking an Armenian revolt, in 571. 
The Byzantines withheld their payment- and in- 
decisive warfare was renewed over Armenia and 
Mesopotamia from 572 to 582. There was a war with 
the Turks in 569-70, and between 575 and 577 a 
Sasanid naval expedition conquered al-Yaman [q.v.] 
from the Ethiopians. The Armenian revolt ended with 
a general amnesty in 578. 

The conflict between the crown and the wuzurgdn 
resurfaced in the reign of Hurmizd IV (579-90), who 
had been designated as successor by his father 
Khusraw I. Hurmizd IV is said to have favoured the 
common people against the wuzurgdn, possibly as a 
basis of support for the crown, killed some 13,000 Per- 
sian notables, executed the Jewish exilarch in 581 and 
closed the schools, and reduced the pay of the army. 
He also suppressed the priestly order of mowbeds; for 
the rest of the Sasanid period and into early Islamic 
times, herbeds are represented as the most important 
priests, although Hurmizd IV resisted their efforts to 
persecute Jews and Christians. In 588 Hephthalite 
subjects of the Western Turks invaded the east, 
reaching Badghls [q.v.] and Harat. Bahram Cubin 
[see bahram], of the Mihran family and spdhbed of the 
North, defeated the Western Turks at Harat in 589, 
crossed the Oxus and defeated the Eastern Turks, and 
was then sent against the Byzantines in Albania, 
where he was defeated. Jealous of Bahrain's populari- 
ty, Hurmizd IV disgraced him on the pretext that he 
held back booty, provoking Bahram to rebel in the 
trans-Caucasus, according to Theophylact, or in the 
east, according to the Syriac Khuzistan chronicle. 
Towards the end of 589 Bahram marched against al- 
Mada'in, where the nobles rebelled early in 590, led 
by Bindoe and Bistam, brothers-in-law of Hurmizd, 

who released the nobles from prison, deposed, blinded 
and killed Hurmizd IV, and enthroned his son 
Khusraw. Bahram defeated them in Adharbaydjan, 
Khusraw fled to the Byzantines in March, Bahram 
entered the capital in the summer of 590 and took the 
throne (Bahram VI, 590-1) with upper-class support, 
including the rich Jews there. He struck coins for two 
years and may have associated his rule with millennial 
expectatations and Arsacid restoration. 

Khusraw sought aid from the Byzantine emperor 
Maurice, who sent him money and two armies for his 
promise to return conquered territory and permanent 
peace. One army marched through Armenia^ rein- 
forced by 12,000 Armenians and awaited in Adhar- 
baydjan by Bindoe with 8,000 Persians. The other 
went through Mesopotamia and defeated Bahram in 
c Irak; Bahram headed for Adharbaydjan, was 
defeated by Bindoe near Lake Urmia, and fled across 
the Oxus to the Turks, where he was assassinated a 
year later. When Khusraw returned to al-Mada'in, 
his general and the Christians killed many of the Jews 
of Mahoza for supporting Bahram. 

Khusraw II (591-628) returned Dara and 
Mayyafarikin [q.v. ] to the Byzantines in 591, and 
ceded Armenia up to Lake Van and Tiflis. He remit- 
ted half the land tax, put one uncle, Bindoe, in charge 
of the administrative bureaux and treasuries, and the 
other, Bistam, over the north-east. Bindoe was soon 
executed, and Bistam rebelled in the north-east, strik- 
ing his own coins at Rayy from about 592 to 596. In 
598 an expedition annexed al-Yaman as a province. 
Khusraw II established his rule over the entire state 
by 601. 

The reign of Khusraw II was the most extreme ex- 
pression of late Sasanid political absolt 
perial ambition. In a letter 
himself "king of kings, mas 
power, lord of peoples, print 
men, good and eternal man an 
ful god among men, most 

cended with the 

, he called 
ter of those who have 
:e of peace, saviour of 

He v 


s the 


is observed in 624 on the inside of a dome on a 
building at Ganzak enthroned in the heavens with the 
sun, moon and stars around him. He is said to have 
had a celestial throne with a canopy of gold and lapis- 
lazuli on which the stars, signs of the zodiac, planets 
and the seven climes were represented. He is also said 
to have been surrounded by over 360 astrologers and 
magicians, whose advice he sought constantly. He 
was called Aparwez (Parwlz [q.v.] "the trium- 
phant"), and the slogan "may he make Iran prosper" 
occured on some of his coins. The rock reliefs in the 
large grotto at Tak-i Bustan are probably his, as are 
the remains of a palace at Kasr-i Shlrin [q. v. ] near 
Dastagerd (Daskara [q.v.]). He amassed a huge royal 
treasure, accumulated from the time of Firuz, 
reorganised the state into 35 administrative districts 
and built 353 fire temples, where 12,000 Imbeds 
recited rituals. 

He eliminated all rival sources of power, probably 
in compensation for insecurity at the beginning of his 
reign. The Jewish exilarchate was suppressed; after 
590 there was no exilarch for the rest of the Sasanid 
period, Jewish schools were closed and the rabbis 
became the de facto leaders of the community. In 596 
he had his own candidate elevated as • Nestorian 
catholicos, but being deceived by a group of physi- 
cians, astrologers and Christian courtiers over the 
elevation of a successor in 604, after the latter died in 
609, he refused to allow the election of a catholicos for 
the rest of his reign. In about 602 the Lakhmids were 

suppressed and their last king, al-Nu c man III b. al- 
Mundhir [q.v.], was executed. After a Sasanid army 
was defeated by the Banu Shayban at Dhu Kar [q. v. ] 
between 604 and 611, al-HTra was put under a marzbdn 
and the desert frontier restored. 

In 602 Maurice was deposed and killed, giving 
Khusraw the opportunity to regain territory ceded to 
the Byzantines by posing as his avenger, starting the 
last and greatest Sasanid-Byzantine war. Between 604 
and 610 Sasanid armies under Shahin, spdhbed of the 
West, and Shahrbaraz conquered Armenia and 
Mesopotamia, and invaded Syria and Anatolia. 
Byzantine weakness during this phase of the war was 
due to their lack of local support and the occupation 
of their forces putting down local disturbances and 
revolts, that destroyed their own manpower in Syria. 
The second phase started with the accession of 
Heraclius in 610; Khusraw now intended to conquer 
the entire Byzantine empire. In 61 1 with Shahln's ar- 
my covering in Anatolia, Shahrbaraz invaded Syria, 
taking Antioch, and Damascus (613). Then marching 
down the coast taking the towns to prevent Byzantine 
reinforcements by sea, he turned inland, conquered 
Galilee and the Jordan valley and besieged Jerusalem 
in May 614, joined by tens of thousands of messianic 
Jews. When Jerusalem fell, tens of thousands of 
Christians were killed, 300 monasteries and churches 
burned, 35,000 captives, mostly craftsmen, deported 
to Sasanid territory along with the Patriarch of 
Jerusalem and the relic of the True Cross, and a 
Bahram fire established there. The Jewish alliance 
broke down in 617, their administration of Jerusalem 
was ended and they were expelled from the city. The 
treatment of Christians in Palestine was moderated by 
the influence of Christians at the Sasanid court; some 
prisoners were freed, money was sent for relief, and 
the rebuilding of churches and monasteries was al- 
lowed. Chalcedonian bishops were expelled from 
Mesopotamia and Syria, and their churches turned 
over to Monophysites. From 616 to 620 Shahrbaraz 
conquered Egypt as far as Ethiopia in the south and 
Libya in the west, and installed a Coptic Patriarch in 
Alexandria. Meanwhile, an Armenian army under 
Sembat Bagratuni defeated Hepthalites north of Lake 
Helmand and looted as far as Balkh [q. v. ]. During this 
phase of the war, the Sasanids were inconsistent in 
responding to local support from Jews and 
Monophysites, and may have played them off against 
each other; they also failed to develop a navy, and 
Khusraw II was unwilling to consider peace while he 

Instead of trying to reconquer lost provinces, the 
Byzantines took the war to the Sasanids in its third 
phase. In 622 and 624 Heraclius invaded Armenia, 
faced by Khusraw II himself in Adharbaydjan. Burn- 
ing towns and villages, Heraclius drove Khusraw II 
out of Ganzak and Shiz. Khusraw II fled to al- c Irak, 
while Heraclius wintered in Albania. In 625 Heraclius 
defeated three Sasanid armies in Adharbaydjan and 
Armenia. With the war going badly, Khusraw II 
turned against the Christians, executed Nestorians in 
upper c Irak, deported Monophysites from Edessa to 
raised taxes and took treasures from churches 

o fina 

e the- 

In its fourth phase, the war escalated in 626; the 
Sasanids allied with the Avars in eastern Europe, 
while the Byzantines allied with the Khazars [q.v] 
north of the Caucasus. The joint Sasanid-Avar siege 
of Constantinople failed in 626, as did the Byzantine- 
Khazar siege of Tiflis. In 627 plague broke out in 
Palestine, and Heraclius invaded Armenia with 
Greek, Laz and Iberian troops, and with Khazar 

cavalry. Descending to the Tigris valley, he defeated 
Sasanid forces at Nineveh, marched down the east 
bank of the Tigris, took and destroyed the royal 
palace and animal preserve at Dastagerd, and took the 
royal treasures. Khusraw II fled to al-Mada'in and set 
up a defensive line along the Nahrawan [q.v.] canal. 
Unable to find a ford due to disastrous flooding of the 
Tigris and Euphrates in the winter of 627-8, Heraclius 
returned to Ganzak, and waited for political 
developments in the capital. Khusraw II suspected 
and feared everyone, arrested all the officers who had 
fled from the Byzantines and ordered them to be ex- 
ecuted, imprisoned thousands of people and executed 
the workmen who failed to close the Tigris breaches. 
Furious at his behaviour and rejection of peace pro- 
posals, in February 628 a group of generals and high 
officials entered the capital, opened the prisons, pro- 
claimed his son Kubadh II king, arrested and im- 
prisoned Khusraw II and insisted that Kubadh ex- 
ecute him, because there could not be two kings at 

Kubadh II reigned for eight months (February to 
September 628), was called Kubadh Flruz ("vic- 
torious") on his coins and was popularly called 
Shlruya/Shiroe. Persians called him "the unjust", 
and he was identified as the Antichrist in apocalyptic 
literature. He was remembered as a parricide and 
fratricide; he killed all of his adult brothers, leaving 
only sisters and children, creating the subsequent 
dynastic problems. During his reign, one-third to one- 
half of the population of c Irak perished from plague, 
probably brought back from Syria by Sasanid armies. 
He also began peace negotiations with the Byzantines 
based on mutual evacuation and the freeing of 
prisoners of war and deportees. He reduced taxes and 
allowed the election of a Nestorian catholicos. 

Kubadh II was succeeded by his seven-year-old 
son, Ardashir III (September 628 to April 630), with 
Mah-Adur-Gushnasp as regent. Shahrbaraz seems to 
have broken with Khusraw II by the end of 626, and 
had refused to recognise Kubadh II or evacuate his 
provinces. In the summer of 629 he negotiated with 
Heraclius on his own, evacuated Egypt and Syria, 
returned to al-Mada'in, purged his enemies and those 
responsible for the death of Khusraw II, and made 
himself regent. He returned the True Cross to 
Heraclius, but was defeated by the Khazars. On 27 
April 630, he killed Ardashir III, made himself king 
and reigned for 42 days until he was killed by his own 
guard on June 9. 

This was followed by an extreme dynastic crisis, 
with eleven rulers in two years. Buran (630-1), 
daughter of Khusraw II and wife of Kubadh II, struck 
coins, built bridges, and completed the peace negotia- 
tions with the Byzantines before being deposed and 
killed in the fall of 631. In the latter part of 632, a 
grandson of Khusraw II, Yazdadjird III [q.v.] (632- 
51), was proclaimed king at Istakhr at the age of six- 
teen or twenty-one and brought to al-Mada'in. 

Yazdadjird III was the last Sasanid monarch. The 
Sasanid position in the Arabian peninsula had already 
been lost; their system of military colonies and tribal 
alliances in al-Yaman, c Uman, and al-Bahrayn col- 
lapsed when defeat by the Byzantines followed by a 
four-year dynastic crisis (628-32) made them unable 
to support their garrisons and Arab proteges. The 
treaty of Hudaybiya [q.v.] in 628 enabled the Muslims 
to form their own alliances, and Sasanid governors at 
San c a' [q.v.] and Hadjar acknowledged the Prophet 
Muhammad and converted to Islam, while Magians 
in al-Bahrayn and coastal c Uman were allowed to pay 
tribute. The raid of Khalid b. al-Walid [q.v.] in 633 

destroyed most of the Sasanid system of fortresses 
along the desert frontier of c Irak, crippled their Arab 
allies there and provoked a major Sasanid effort to 
restore the border. But they were unable to follow up 
their victory over the Muslims at the Battle of the 
Bridge in 634 because of factional conflict at al- 
Mada'in. The Muslim victory at al-Kadisiyya [q.v.] 
in 636 resulted in the loss of c Irak to the Sasanids; 
Yazdadjird fled to Hulwan [q.v.] and then to Rayy. 
The fall of c Irak affected the subsequent conflict 
because the Sasanids had lost the heart of their state: 
the administrative centre at al-Mada'in, the tax base, 
that amounted to one-third of the total revenue, the 
royal treasure, substantial military forces and the 
leadership of many nobles. The Muslims now held 
these resources, assisted by former Sasanid soldiers 
and officials who defected to them. The caliph c Umar 
'. [q.v.] had intended expansion to stop there, but at- 
tacks on lower c Irak by Sasanid forces in Khuzistan 
provoked the Muslim conquest of that province from 
639 to 642. Yazdadjird III raised a major army and 
it to Nihawand [q.v.] in order to block any 
lim advance and possibly to retake c Irak. The 
defeat of this army by combined Kufan and Basran 
forces in 642 was a second military disaster for the 
Sasanids; it secured Khuzistan and c Irak for the 
Muslims, ended organised resistance in the Djibal 
[q.v.] and opened the Iranian plateau to the Muslims. 
Yazdadjird III fled to Isfahan and then to Istakhr, 
where he tried to organise the defence of Fars. But the 
Basran army conquered Fars in 649-50; Yazdadjird 
III fled to Kirman and Sistan, pursued by Muslim 
forces, and arrived at Marw. Resenting Yazdadjird 
Ill's financial demands, the marzban of Marw allied 
with Nizak Tarkhan [ q . v _] the Hephthalite ruler of 
Badghfs, to defeat Yazdadjird's followers. Yazdadjird 
III fled from Marw, and was killed by a miller nearby 
in 451. His son Flruz took refuge in T'ang China, but 
he and his son Narseh were unsuccessful in getting 
Chinese help to restore their dynasty. The fall of the 
Sasanids did not mean the Muslim conquest of Persia, 
which took several more decades [see Iran. v. 
History). The Sasanid Book of Kings (Xwaday ndmag) 
achieved its final form in the reign of Yazdadjird III, 
and, because he had no successor, his regnal years 
continued from 632 as the era of Yazdadjird. Magians 
took that year as the end of the millennium of 
Zoroaster and the beginning of the millennium of 
Oshedar. . 

The significance of Sasanid history lies in providing 
an example of a late antique state and society that 
broadens the understanding of that period, in the 
development of monarchic and religious institutions, 
and the formation of religious communities, that 
created precedents for religious groups as political 
minorities. The Sasanids left a legacy of royal ab- 
solutism and bureaucratic administration, and 
Sasanid motifs continued in the art and architecture of 
the Islamic period and spread to the east and west. 
Branches of the dynasty survived long afterward as 
local rulers. The Bawandids (Bawand [q.v.]) of 
Tabaristan [q.v.] claimed descent from Kawus, the 
son of Kubadh I, and ruled until 750/1349. The 
Dabawayhids (DSbuya [q.v.]) claimed descent from 
Djamasp, the son of Firuz, and ruled in Tabaristan 
and Gllan in the 7th and 8th centuries. The 
Badusbanids [q.v.] claimed descent from them, and 
branches of this dynasty lasted until 975/1567 and 

Bibliography: The literature on subjects related 
to the Sasanids is huge; only an updated selection is 

tory bibls. are given by A. Chnstensen, L'Iran 
sous les Sassanides 2 , Copenhagen 1944, now 
superseded by the Cambridge hist, of Iran, m, Cam- 
bridge 1983; J. Neusner, A history of the Jews in 
Babylonia, ii-v, Leiden 1966-70; articles in La Persia 
nel Medwevo, Rome 1971; and RN Frye, The 
heritage of Persia, London 1962, and The history of An- 
cient Iran, Munich 1984. Orientals J Duchesne- 
Guillemin emerito oblala, Acta Iranica 23, Deuxieme 
serie, Hommages et Opera Minora, ix, Leiden 



i the 5 

period. See also the proceedings of a symposium at 
Turin, 1985, Common ground and regional features of the 
Parthian and Sasanian world, in Mesopotamia, x\n 
(1987), 17-21; F. Rahimi-Laridjani, Die Entwuklung 
der Bewasserungslandwirtschaft im Iran bis in sasamdisch- 
fruhislamische Zeil, Wiesbaden 1988; G. Gnoli, The 
idea of Iran: an essay on its origin, Rome 1989; R. 
Brody, fudaism in the Sasanian empire: a case study in 
religious coexistence, in S. Shaked and A. Netzer, 
Irano-fudaica, ii, Jerusalem 1990, 52-62; K. Schipp- 
mann, Grundziige der Geschichte des Sasanidischen 
Reiches, Darmstadt 1990. For Armenia, see N. Gar- 

a betwe 


w Sasan 

London 1985. 

For inscriptions, see the Corpus Inscriptionum 
Iranicarum, III, Pahlavi Inscriptions, London 1955-; 
Ph. Gignoux, Glossaire des Inscriptions pehlevies et par- 
thes, London 1972, with a complete bibl. of previous 
publications; M. Back, Die sassanidischen Staats- 
inschriften, Leiden 1978. Specific inscriptions are 
noted below according to period. 

On coins, see F.D.J. Paruck, Sassanian coins, 
Bombay 1924, repr. 1976; R. Gobi, Sasanian 
numismatics, Brunswick 1971; M. Mitchener, The 
mint organization of the Sassanian Empire, in Numismatic 
Circular, lxxxvi (1986); R. Curiel and R. Gyselen, 
Monnaies des fouilles de Bishdpur, in Stir xvi (1987), 7- 
43. Other works on coins are given below by 


ireated by Gobi, Der 
iwick 1973; idem, Die 
eiman, Berlin 1976; 

sasanidische Siegelkano 
Tonbullen mm Tac 
Gyselen, Une classifii 
forme, in Stir, v (1976), 139-46; eadem, Le reemploi 
de sceaux a I'epoque sassanide, in ibid., xx (1991), 203- 
10; D. Huff, Technological observations on clay bullae 
from Takht-i Suleiman, in Mesopotamia, xxii (1987), 
367-90; Gignoux and Gyselen, Bulles et sceaux 
sassanides de dioerses collections, Paris 1987; eidem, 
Sceaux defemmes a I'epoque sassanide, in L. De Meyer 
and E. Haerinck (eds.), Archaeologia Iranica et Orien- 
tals, Ghent 1989, 877-96. See also S. Shaked, 
Jewish and Christian seals of the Sasanian period, in M. 
Rosen-Ayalon Studies in memory of Gaston Wiet, 
Jerusalem 1977, 17-31; and J. A. Lerner, Christian 
seals of the Sasanian period, Istanbul 1977; corrections 
in H. Humbach, ZDMG, cxxix (1979), 189-90; 
critique by Gignoux, Sceaux chretiens d'epoque 
sasanide, in Iranica Antiqua, xv (1980), 297-314. 

For institutions, see K. Erdmann, Die Ent- 
wicklung der sasanidischen Krone, in Ars Islamica, xv-xvi 
(1951), 87-123; F. Altheim and R. Stiehl, Ein 
asiatischer Staal, Wiesbaden 1954; Frye, Notes on the 
early Sassanian state and church, in Studi Orientalistici in 
onoredi Giorgio Levi Delia Vida, Rome 1956, 314-35; 
idem, The Sasanian system of walls for defense, in Studies 
in Memory of Gaston Wiet, Jerusalem 1977, 7-15; D. 
Metzler, Ziele und Formen kbniglicher Innenpolitik im 
oorislamischen Iran, Munster 1977; D.M. Goodblatt, 
The poll tax in Sasanian Babylonia, in JESHO, xxii 
(1979), 233-94; K. Mosig-Walburg, Die fruhen 

schen Komge als Vert 

reter und Forderer a 

trischen Religion, Fr 

nkfurt a.M. 198 

stration in sasanidiscf 

Vberbluk, m H Koch 

and D N MacKenz 

Aihamemdinzeil und ihr Fortleben, Berlin 1983; idem, 

ban, mJSAI, iv (1984), 1-29, idem, Le spahbed, des 
Sassanides a VIslam, in ibid , xm (1990), 1-14; W. 
Sundermann, Ke Cihr az yazdan Zur Titular der 
Sasamdenkonige, in ArO, lvi (1988), 338-40; Gyselen, 
Les donnees de_ geographic administrative dans le 
"Sahrestdmhi-i Erin", in Stir, xvn (1988), 191-206; 
eadem, La geographic administrative de I'Empire 
sassanide. Pans 1989, and Shaked, Administrative 
Functions of priests in the Sasanian period, m Gnoli and 
A. Panaino, Proceedings of the First European Conference 
of Iranian Studies, Rome 1990. 

For Magian religion, one can start with K. 
Erdmann, Das iramsche Feuerheiligtum, Leipzig 1941; 
J. de Menasce, Feux et fondations pieuses dans le droit 
sassanide, Paris 1964; Humbach, Atur Gusnasp und 
Takht i Suleiman, in Festschrift fur Wilhelm Eilers, 
Wiesbaden 1967, 189-91; K. Schippmann, Die 
iranischen Feuerheiliglurner , Berlin 1971; J. Duchesne- 
Guillemin, The religion of ancient Iran, Bombay 1973; 
idem, Art and religion under the Sasanians, in Gignoux 
and A. Tafazzoli, Memorial Jean de Menasce, 
Louvain-Tehran 1974, 147-54; M. Boyce, A history 
of Zoroastnanism, Leiden-Cologne 1975; eadem, 
Iconoclasm among Zoroastrians , in Neusner (ed.), 
Christianity, Judaism and other Greco-Roman culls. 
Studies for M. Smith at sixty, Leiden 1975. More 
recently, see Gignoux, Pour une esquisse des fonctions 
religieuses sous les Sasamdes, in JSAI, vii (1986), 93- 
108; idem, Vn categone de mages a la fin de I'epoque 
sasanide: les mogveh, in JSAI, ix (1987); J.R. 
Russell, Zoroastnanism in Armenia, Cambridge, 
Mass. 1987; Gyselen, Note de glyplique sassanide: quel- 
ques elements d'iconographie religieuse, in F. Vallat 
(ed.), Contribution a I'Histoire de I' Iran. Melanges offerts 
a Jean Perrot, Paris 1990, 253-67; Boyce, Some further 
reflections on Zurvanism, in eadem and G. Windfuhr 
(eds.), Iranica varia: papers in honor of Professor Ehsan 
Yarshaler, Acta Iranica 30, Leiden 1990, 20-9; A.S. 
Melikian-Chirvani, Le livre des wis, miroir du destin. 
II Takht-e Soleymdn et la symbolique du Shah-Name, in 
< (1991), 33-148 

On a 


^.U. Pope (ed.), Survey of Persian art, i, iv, Lon- 
don 1938, repr. London 1964-5; K. Erdmann, Die 
Kunsl bans zur Zeit der Sasaniden, Berlin 1943, Mainz 
1969; V.G. Lukonin, Persia II, Geneva 1967; W. 

Hinz, Allira 

he Funde und Forsc 

i, Berli 


i and K. Horiuchi, Taq-t Bustan, Tokyo 
1972; J. A. Neely, Sassanian and early Islamic water- 
control and irrigation systems on the Deh Luran plain, in 
Th.E. Downing and McG. Gibson (eds.), Irriga- 
tion's Impact on Society, Tucson 1974, 21-42; S.A. 
Matheson, Persia: an archaeological guide, London 
1976; P.R.S. Moorey, Kish excavations 1923-1933, 
Oxford 1978, repr. 1979; P. Harper, Royal hunter: 
art of the Sasanian empire, New York 1978; L. Vanden 
Berghe, Bibliographic analylique de I'arche'ologie de ITran 
ancien, Leiden 1979; G. Herrmann and R. Howell, 
The Sasanian rock reliefs at Bishapur, Berlin 1981; R. 
Boucharlat and J. Salles, The history and archaeology of 
the Gulf from the fifth century B.C. to the seventh century 
A.D.: a review of the evidence, in Procs. of the Seminar for 
Arabian Studies, xi (1981), 65-94; K.M. Trinkaus, 
Pottery from the Damghan plain, Iran: chronology and 
variability from the Parthian to the early Islamic periods, in 


Stir, xiv (1986), 23-88; Harper, Parthian andSasanim 

Mesopotamia, xxii (1987), 341-55; Boucharlat, Suse t 
I 'epoque sasanide. Une capitate prestigieuse devenue ville de 
province, in ibid. , xxii (1987), 357-66; and R. Wenke 
and N. Pyne, Some issues in the analysis of Sasanian 
Iran, in V allat (ed.), Melanges ... Jean Perrot, 235-51. 

For the third century, Shapur I's inscription at 
Naksh-i Rustam (SKZ) was published by E. 
Honigmann and A. Maricq, Researches sur les Res 
Gestae Divi Saporis, Brussels 1953; and Maricq, 
Classica et orientalia 5. Res gestae Divi Saporis, in his 
Classica et orientalia, extrait de Syria 1955-62, Paris 
1965, 37-101; his inscription at BTshapur by R. 
Ghirshman, Inscription du monument de Chdpour I", in 
Revue des Arts Asiatiques, x (1937), 123-29; and at 
Hadjiabad by MacKenzie, Shapur's shooting, in 
BSOAS, xli (1978), 499-511. On the establishment 
of a fire temple at the beginning of Shapur I's reign, 
see Gignoux, D'Abnun a Mahdn: etude de deux inscrip- 
tions sassanides, in Stir, xx (1991), 9-22. For coins, 
see S.D. Loginov and A.B. Nikitin, Sasanian coins of 
the third century from Merv, in Mesopotamia, xxviii 
(1993), 225-46. For the graffiti at Dura, see B. 
Geiger, The Middle Iranian texts, in A.R. Bellinger et 
al. (eds.), The Excavations at Dura-Europos. Final report 
VII, Part I, New Haven 1956, 283-317. On the 
death of Mini, see J. P. Asmussen, Manichaean 
literature, Delmar, N.Y. 1977. For Kirdir's in- 
scriptions, see Frye, The Middle Persian inscription 
of Kartir at Naqs-i Rajab, in Indo- Iranian Journal, viii 
(1964-5), 211-25; M.-L. Chaumont, L 'Inscription de 
Kartir a la "Kaaba de Zoroastre" , in J A (I960); 
Gignoux, L 'Inscription de Kartir a Sar Mashad, in ibid. 
(1968), 387-418; idem, Etudes des variantes lextuelles 
des inscriptions de Kirdir, in Le Museon, lxxxvi (1973), 
193-216; idem, Les quatre inscriptions du mage Kirdir. 
Textes et concordances, Paris 1991; Herrmann, 
MacKenzie and R.H. Caldecott, Iranische 
Denkmaler, Reihe II, Iranische Felsreliefs. Lieferung 
13,1, Naqsh-i Rustam 6. The Roman victory of Shapur 
I and the bust and inscription of Kerdir, Berlin 1989. See 
also F. Grenet, Observations sur le litres de Kirdir, in 
Stir, xix (1990), 87-94. The best recent treatment of 
Narseh's inscription is by Humbach and P.O. 
Skjaerve, The Sassanian inscription of Paikuli, 
Wiesbaden 1978-83. Political events from the 
Roman point of view are given by Herodian, 
History of the Empire from the time of Marcus Aurelius, tr. 
C.R. Whittaker, Cambridge, Mass. 1970; Cassius 
Dio, Roman History, tr. E. Cary, Cambridge, Mass. 
1955; The Scriptores Historiae Augustae, ed. and tr. D. 
Magie, London-Cambridge, Mass. 1921-32. See 
also A. Rosger, Die Darstellung des Per serf eldzugs des 
Severus Alexander in der Historia Augusta, in Bonner 
Historia-Agusta-Colloquium 1975176, Bonn 1978. In- 
scriptions and exerpts from contemporary authors 
are translated in J. Gage, La montee der Sassanides et 
I'heure de Palmyre, Paris 1964. 

General works on the third century include 
Chaumont, Les Sassanides et la christianisation de V Em- 
pire iranien au III' siecle de notre ere, in RHR, clxiv 
(1964), 165-202; eadem, La christianisation del' empire 
iranien des origines aux grands persecutions du 4' siecle, 
Louvain 1988; Lukonin, Iran v III veke, Moscow 
1979. For Sasanid origins, see Chaumont, Le 
culte de Anahitd a Stakhr et les premiers Sassanides, in 
RHR, cliii (1958), 154-75. Works on Ardashir I in- 
clude V.G. Loukonin, Monnaie d'Ardachir I et I'Art 
officiel sassanide, in Iranica Antiqua, viii (1968), 106- 
17; P. Schmitt, Artaxerxes, Ardasir und Verwandte, in 
Incontri Linguistici, v (1979), 61-72; V. Fiorani 

Piacentini, Ardashir i Papakan and the wars against the 
Arabs. Working hypothesis of the Sasanian hold of the 
Gulf, in Procs. of the Eighteenth Seminar for Arabian 
Studies, London 1985, 57-77; D. Potter, Alexander 
Severus and Ardashir, in Mesopotamia, xxii (1987), 147- 
57. On the fall of Hatra, see Maricq, Les dernieres 
anne'es de Hatra, in Syria, xxxiv (1957); Chaumont, A 
propos de la chute de Hatra et du couronnement de Shapur 
I", in Acta Antiqua Scient. Acad. Hungarica, xxvii 
(1979 [1981]); J. Wiesehofer, Die Anfdnge der 
sassanidischen Welipoliiik und der Vntergang Hatras, in 
Klio, lxiv (1982), 437-47. On Shapur I, see also 
Sunderman, Shapur's coronation. The evidence from the 
Cologne Mani Codex reconsidered and compared with other 
texts, in Aspects of Iranian culture, in honor of Richard 
Nelson Frye, Bulletin of the Asia Institute, iv/2 (1990), 
295-9; and K. Mosig-Walburg, Shapur I. "Konigvon 
Iran" - Faktum oder Irrlum?, in Schweizerische 
sche Rundschau, lxix (1990), 103-26. For 


: third ( 


mlpturesofBahramll, mJRAS(1970), 165-71;J.K. 
Choksy, A Sasanian monarch, his queen, crown prince 
and dieties. The coinage of Wahram II, in American Jnal. 
of Numismatics, i (1989), 117-37; and W. Seston, Le 
roi sassanide Narses, les arabes, et le Manicheisme, in 
Melanges syriens offerts a M. Rene Dussaud, i, Paris 
1939, 227-34. J. Cribb, Numismatic evidence for 
Kushano-Sasanian chronology, in Stir, xix (1990), 151- 
93, argues for the third and early fourth century, as 
does M.L. Carter, Early Sasanian and Kushano- 
Sasanian coinage from Merv, in Aspects of Iranian 
cultures, 11-16. 

Primary sources for the fourth century in- 
clude R. Frye, The Persepolis Middle Persian inscrip- 
tions from the time of Shapur II, in AO, xxx (1966), 83- 
93; Ammianus Marcellinus, Rerum Geslarum Libri 
qui supersunt, ed. and tr. J. Rolfe, Cambridge, 
Mass. 1950. On the latter, see also K. Rosen, Stu- 
dien zur Darstellungskunst und Glaubwurdigkeit des Am- 
mianus Marcellinus, Bonn 1970. Standard editions of 
the Acts of the Persian martyrs are by E. Assemani, 
Acta martyrum orientalium el occidentalium, Rome 1748, 
and P. Bedjan, Acta martyrum et sanctorum, Paris 
1890-7, with translations by G. Hoffman, Auszuge 
aus syrischen Akten persischer Mdrtyrer, Leipzig 1880, 
and O. Braun, Ausgewdhlte Akten persischer Mdrtyrer, 
Kempten-Munich 1915. See also H. Delahaye, Les 
versions greques des actes des martyrs persons sous Sapor II, 
in PO, ii/4 (1906), 401-560; and P. Devos, Les mar- 
tyres persons a travers leurs actes syriaques, in La Persia e 
il mondo Greco-Romano, Rome 1966, 213-25. The 
Armenian account of Faustus of Byzantium is now 
translated by Garsoian, The epic histories attributed to 
P'-awstos Buzand (Buzandaran PatmutHwnk c ), Cam- 
bridge, Mass. 1989. For the age of Shapur II, see 
Neusner, Aphrahat and Judaism. The Christian-Jewish 
argument in fourth-century Iran, Leiden 1970; M.I. 
Mochiri, Evenements historiques, localisations geographi- 
ques et indice chronologique attestes par les monnaies de 
Shapur II, in Iran, xxvii (1989), 39-49; idem, 
Titulature de ShdpUr II, in Iran, xxviii (1990), 13-22; 
Gignoux, Les nouvelles monnaies de Shapur II, in Stir, 
xix (1990), 195-204; S.D. Loginov and A.B. 
Nikitin, Coins of Shapur II from Merv, in Mesopotamia, 
xxviii (1993), 247-70; Gyselen and H. Gasche, Suse 
et Ivan-e Kerkha, capital provinciate d'Eran-xwarrah- 
Sapur. Note de geographie historique sassanide, in Stir, 
xxiii (1994), 19-35. On Shapur III, see Chau- 
mont, A propos d'un edit de paix religieuse d'epoque 
sassanide, in Melanges d'Histoire des Religions offerts a 
H.C Puech, Paris 1974, 71-80. 

In the fifth century, for Mihr-Narseh see W.B. 


Henning, The inscription of Flruzabdd (Pahlavi inscrip- 
tion at the ruins of the Sassanian bridge), in Asia Major, 
n.s. iv/1 (1954), 98-102. For Christian synods 
from the 5th century onwards, see J.B. Chabot (ed. 
and tr.), Synodicon Orientate, Paris 1902. The martyrs 
of Kirkuk are in the 6th-century History of Karka de- 
Bet Selok, ed. P. Bedjan, in Acta martyrum et sanctorum, 
ii, 507-37, tr. G. Hoffmann, Auszuge, 43-60. The 
Armenian accounts by Elise Vardapet, Vasn Var- 
danac ew Hayoc Paterazmin, ed. E. Ter-Minasean, 
Erevan 1957, and Lazar of P c arp, PatuitHun Hayoc, 
ed. Ter-Mkrtzean, Tifiis 1904, are tr. in V. 
Langlois, Collection des histoires anciens de moderns 
d'Armenie, ii, Paris 1869. On the coins of Firuz, 
see W. Szaivert, Die Munzpragung des Sasanidenkbnigs 
Peroz. Versuch einer historischen Interpretation, in Litterae 
Numismaticae Vindobonensis , iii (1987), 157-68. For 
the early Nestorians, see A. Voobus, History of 
the School of Nisibis, Louvain 1965. 

On Kubadh and Mazdak, in addition to the 
literature cited in the article mazdak, see O. Klima, 
Beitrdge zur Geschichte des Mazdakismus, Prague 1977; 
H. Gaube, Mazdak: historical reality or invention?, in 
Stir, xi (1982), 1 1 1-122; Tafazzoli, Observations sur le 
soi-disant Mazdak-Namag, in Acta Iranica, xxiii (1984). 
For a thought-provoking reassessment, see P. 
Crone, Kavad's heresy and Mazdak's revolt, in Iran, 
xxix (1991), 21-42. 

The most important Byzantine sources for 
the 6th century include Procopius, History of the 
Wars, ed. and tr. H.B. Dewing, London 1961; 
Agathias of Myrina, Agathiae Myrinaei Historiarum 
libri quinque, ed. R. Keydell, Berlin 1967, tr. J. 
Frendo, The histories! Agathias, Berlin-New York 
1975, which has the earliest surviving account of 
Sasanid dynastic history. For an assessment of the 
latter, see Averil Cameron, Agathias, Oxford 1970. 
The late Sasanid lawbook is edited by A.G. 
Perikhanian, Matigan e hazar datestan, Erevan 1973; 
and M. Macuch, Das sasanidische Rechtsbuch 
"Mdtakddn i Hazar Ddtistdn" , Wiesbaden 1981. For 
late Sasanid literature, see D. Sanjana, Kdr- 
ndmak i Ardashir i Pdpakan, Bombay 1896, tr. Th. 
Noldeke, in Bezzenberger's Beitrdge zur Kunde der in- 
dogermamschen Sprachen, iv (1879), 22-69; E.K. An- 
tia, Kdrnamak-i Artakhshir Pdpakdn, Bombay 1900; 
Boyce, The Letter of Tansar, Rome 1968; and M. 
Grignaschi, Quelques specimens de la litterature sassanide 
conserves dans les bibliotheques d' Istanbul, in J A, ccliv 
(1966), 1-142, which includes Arabic translations of 
the Testament of Ardashir and the Kdr-ndmag of 
Anushirwan. See also D. Gutas, Paul the Persian o> 
the classification of the parts of Aristotle's philosophy: < 
milestone between Alexandria and Bagdad, in Isl., h 
(1985), 231-67. 

Chobin and U 
Asia Institute 

early career of Kh, 
ii (1989), 77-88. 
Hint of the Sasanid-: 

II, in Bull, ofth 




e Theophyla 

Historiae, ed. C. de Boor, Stuttgart 1972, tr. M. and 
M. Whitby, The History of Theophylact Simocatta, Ox- 
ford 1986; and the Khuzistan chronicle, ed. and tr. 
I. Guidi, Chronica minora I, CSCO, Scr. Syri\, 16-39, 
and 2, 13-32, Louvain 1955, tr. Noldeke, Die von 
Guidi herausgegebene syrische Chronik, in SB Wiener Ak. 
Wiss., cxxviii (1893), 1-48. For this period, see also 
M. Higgins, The Persian War of the Emperor Maurice, 
Washington, D.C. 1939; P. Peeters, Les ex-voto de 
Khosrau Aparwez a Sergiopolis, in Analecta Bollandiana, 
lxv (1947), 5-56; H.P. L'Orange, Studies on the 
iconography of cosmic kingship in the Ancient World, Oslo 
1953; K. Czegledy, Bahram Cobin and the Persian 
apocalyptic literature, in AO, viii (1958), 21-43; D. 
Frendo, Theophylact Simocatta on the revolt of Bahram 

and guide tc 

Stratos, Byzantium in the seventh century, tr. M. 
Ogilvie-Grant, i, Amsterdam 1968. For the late 
Sasanid Book of Kings, see A.S. Shahbazi, On the 
X'addy-ndmag, in Iranica varia: papers in honor of Pro- 
fessor Ehsan Yarshater. The most important post- 
Sasanid Middle Persian literature with infor- 
mation on the Sasanid period includes_ J. 
Markwart, A catalogue of the provincial capitals of Erdn- 
shahr, Rome 1931; D.M. Madan, The complete text of 
the Pahlavi Dinkart (DKM), Bombay 1911; B.T. 
Anklesaria, Zand-Akasih, Iranian or Greater Bundahisn, 
Bombay 1956. Much historical and legendary ma- 
terial on the Sasanids survived translated into 
Arabic and Persian, but this literature must be used 
critically. The most informative works include the 
anonymous Nihdyat al-arab fi akhbdr al-Furs wa 7- 
<Arab, mss. Cambridge Qq225, Cairo taMkh 4505, 
summarised by E.G. Browne in Some account of the 
Arabic work entitled "Nihdyatu 'l-irab fi akhbdr 'l-Furs 
wa 'l-'-Arab, " particularly of that part which treats of the 
Persian Empire, in JRAS (1900), 195-259; Tha'alibi, 
Ghurar akhbdr muluk al-Furs wa-siyaruhum, ed. and tr. 
H. Zotenberg, Histoire des rois des Perses, Paris 1900; 
ft.-Djahiz, K. al-Tddx ft akhldk al-muluk, Cairo 
1322/1914, tr. Ch. Pellat, Le livre de la couronne, 
Paris 1954; DInawari, K. al- Akhbdr al-tiwdl, Leiden 
1912; Ya c kubl, Ta^rikh, Leiden 1883; Taban, 
Sasanid seetion tr. Noldeke, Geschichte der Perser and 
Araber zur Zeit der Sasaniden, Leiden 1879; Isfahan!, 
7VnM sini muluk al-ard wa 'l-anbiya\ Beirut 1961; 
Mas'udl, Murudx, Firdawsl, Shdh-ndma, Moscow 
1963-71; the anonymous Chronicle of SiHrt, in PO, 
iv/3, 219-313, ii, 241-334, vii/2, 99-203, xiii/4, 437- 
639. The late Sasanid period is also covered in 
Greek by Theophanes, Chronographia, ed. C. de 
Boor, Leipzig 1883-5, tr. H. Turtledove, The 
Chronicle of Theophanes, Philadelphia 1982; and in 
Syriac by Michael the Syrian, ed. and tr. Chabot, 
Chronique de Michel le Syrien, Paris 1899-1910, repr. 
Brussels 1963, both of which are based on earlier 
sources. On the Muslim conquest of Persia, see 
Baladhuri, tr. P.K. Hitti, The origins of the Islamic 
state, New York 1916, repr. Beirut 1966. For addi- 
tional bibliography, see M. Morony, "-Arab. ii. Arab 
conquest of Iran, in EIr, ii (1987), 203-10. 

(M. Morony) 
SASARAM [see sahsaram]. 
SATALIA [see antalya]. 

SATGA'ON, Saptagrama in Sanskrit, a famous 
medieval port city and administrative centre in 
southwestern Bengal. Located at the junction of 
the rivers Bhagirathi and Saraswati and adjacent to 
both Triveni— a holy place to the Hindus— and Chota 
Pandu'a (see pandu'a], the city existed long before its 
conquest by a famous Muslim army commander 
Zafar Khan GhazI during the reign of Sultan 
Kaykawus Shah (689-700/1290-1301). A thriving port 
city and commercial place during the Sultanate 
period, Satga'on also became an important Muslim 
cultural and educational centre where large number of 
(such as the Madrasa Dar al- 

Khayrat dated 713/1313) i 

governors such as Tarbiyat Khar 
Barbak Shah (later Sultan Bart 
and Ikrar Khan in 860/1455. 

Among its Muslim architectural remains is a 
mosque from the reign of Sultan Nusrat Shah dated 

: built. Epigraphic 
mes of its celebrated 
i in 861/1457, Malik 
k Shah) in 860/1455 


936/1529 built by a saint Sayyid Djamal al-DIn 
Husayn b. Sayyid Fakhr al-DIn, an immigrant from 
the Caspian coast town of Amul (see Shamsuddin 
Ahmad, Inscriptions of Bengal, v, Rajshahi 1960, 24-7, 
28-9, 56-7, 68-70). Satga'on was an important mint 
town since the beginning of 8th/14th century. The 
earliest coin discovered so far is dated 729/1328, while 
the latest is 957/1550. 

The first European contacts with Satga'on date 
from the 1530s when Portuguese ships started using 
its port. Unfortunately, due to the violence often prac- 
tised by the Portuguese in the area, the population of 
this grand city — once called Porto Piqueno or the little 
Heaven by the early Europeans — started dwindling. 
The final ruin of Satga-'on seems to have been brought 
about by the silting of the river Saraswati, as the port 
almost lost its navigability towards the end of 
the 10th/16th century. 

Bibliography: Rakhal Das Bandyopadhyaya, 

Saptagrama or Sdtgdnw, in JASB, v/7 (July 1990), 

245-62; D.G. Crawford, Satgaon and Tribeni, in 

Bengal Past & Present, iii (Jan. -April 1909), 18-26. 
(Muhammad Yusuf Siddiq) 

SATIH b. RABI C A, a legendary diviner (kahin) 
of pre-Islamic Arabia, whom tradition connects with 
the beginnings of Islam; in reality, we are dealing here 
with a quite mythical personage like the other kahin in 
whose company he appears in most stories, Shikk al- 
Sa c bi, who is simply the humanisation of a 
demoniacal monster in appearance like a man cut in 
two {shikk al-insan: cf. van Vloten, in WZKM, vii 
[1893], 180-1, and shikk). Satlh, whose name means 
"flattened on the ground and unable to rise on ac- 
count of the weakness of his limbs" (Lisan al-'-Arab 1 , 
iii, 312), is described as a monster without bones or 
muscles; he had no head but a human face in the cen- 
tre of his chest; he lay on the ground, on a bed of 
leaves and palm branches, and when he had to change 
his position "they rolled him up like a carpet"; only 
when he was irritated or inspired did he inflate himself 
and stand up. His close resemblance to Shikk is accen- 
tuated by legend which makes them both be born 
without the intervention of a father in the night before 
the death of the kdhina Turayfa (the wife of c Amr 
Muzaykiya', ancestor of the tribe of this name, who 
is said to have foretold the catastrophe of the breaking 
of the dam of Ma^rib in the Yemen). She is said before 
dying to have made the two newborn monsters come 
to her and, after spitting in their months (the classic 
method of transmitting magic power), declared them 
her successors in the art of kihdna. 

In spite of these characteristically mythical features, 
Arab genealogical tradition has not refused to give 
Satlh a place in its system, but gives him a name and 
a paternity (Rabl* b. Rabf'a b. Mas c fld b. Mazin b. 
Dhi-'b), which connect him with the Ghassanid branch 
of the tribe of Azd (just as it connects Shikk with the 
Banu Sa c b, a branch of the Banu Badjila) and more 
precisely with the Banu Dhi-'b (Ibn Durayd, Ishtikdk, 
286; Wustenfeld, Genealog. Tabellen, 11, 16; according 
to others, the Banu Dhi-'b belonged to the c Abd al- 
Kays, a tribe belonging to the Rabf'a group); there 
even seems to have been in historic times an Azd clan 
claiming descent from Satlh (Abu Hatim al-Sidjistanl, 
Kitdb al-Mu c ammarin, 3, in Goldziher, Abhandl zur 
arab. Philologie, Leiden 1896-9, ii). 

Among the legends associated with the name of 
Satlh some are connected with the pre-history of the 
Arabs and represent Satlh as acting as a diviner and 
judge (hakarri) without any regard for history or chron- 
ology, being totally fictitious; sometimes we find him 
dividing among the sons of Nizar (Mudar, Rabi'a, 

Iyad and Anmar) their father's estate (%d, ii, 46, 4th 
ed., ii, 39); sometimes we find him consulted with 
Shikk by al-Zarib al- c AdwanI (Wustenfeld, Gen. 
Tabellen, D, 13) regarding the real position of Kasi, 
the ancestor of the Thakif, to whom al-Zarib had been 
forced to promise his daughter in marriage (Aghani, 1 
ii, 75). In al-Ya c kubI(ra>rSA, i, 288-90), it is he who 
decides the difference that arose between c Abd al- 
Muttalib, the Prophet's grandfather, and the two 
Kaysl tribes al-Kilab and al-Ribab, regarding the 
ownership of the well of Dhu '1-Harm discovered by 
the former in the vicinity of al-Ta^if; but the parallel 
versions of the same story either do not mention the 
name of the arbitrator or give him that of another 
kahin, Salama b. Abl Hayya al-Kuda c I (al-Maydanl, 
Amthal, ed. 1284, i, 36 = ed. 1310, i, 30; Yakut, iv, 
629; LW, xiii, 283). 

Two other legends, on the other hand, have a com- 
pletely Islamic stamp; according to the first, given by 
Ibn Ishak, who does not give his sources, Satlh 
consulted — as always, with Shikk — by the Lakhmid 
chief Rabija b. Nasr regarding a dream which had 
frightened him, reveals to him that South Arabia will 
be invaded by the Abyssinians and that after the ex- 
pulsion of the latter and the brief dominion of the Per- 
sians it will be conquered by a Prophet (Muhammad); 
as a result of the oracle, Rabija b. Nasr sends his son 
c Amr at the head of the tribe to the king of Persia who 
settles them at al-HIra; this is the "South Arabian" 
version of the foundation of the Lakhmid dynasty (cf. 
G. Rothstein, Die Dynastie der Lahmiden in al-Hlra, 
Berlin 1899, 39). 

The second and most widely disseminated legend 
goes back to a certain Hani 3 al-Makhzuml. who is 
said to have lived to the age of 150 years and about 
whom Muslim historiographical tradition knows 
nothing (see Ibn Hadjar, Isdba, vi, 279, no. 8,929). It 
forms part of the cycle of the aHam al-nubuwwa, that is, 
of the miraculous signs which confirm the truth of the 
prophetic mission of Muhammad. In the night when 
the latter was born, remarkable phenomena occurred 
throughout the kingdom of Persia. The king (Kisra 
Anushirwan), not being able to get an explanation 
from his magicians, asked the king of al-HIra, al- 
Nu c man b. al-Mundhir (an anachronism!), to send 
him someone who could explain it. Al-Nu c man sent 
c Abd al-MasIh b. Bukayla al-Ghassanl (on whom see 
al-Sidjistanl, Kitdb al-Mu^ammarin, 38; Caetani, Annali 
dell' Islam, ii, 935, 12 A.H., § 165, iv, 657, 21 A.H., 
§ 328), who, not being able to explain these marvels 
himself, went to Satlh, his maternal uncle, who lived 
in the desert. He found him at the point of death, and 
his appeal was unanswered; only after his nephew had 
addressed him in verse, did the kahin predict to him 
the coming fall of the Persian Empire and its conquest 
by the Arabs, etc. Having delivered this oracle, his 
uncle Satlh died. 

Satlh claimed to receive his knowledge of the future 
from a familiar spirit (ra^i, see kahin), who had 
overheard the conversation of God with Moses on 
Mount Sinai and had revealed part of it to him. Here 
we see the influence of the Kur'anic passage (LXXII, 
1) about the djinn who overhear God's utterances. 

The calculations of the Arab' historians on the age 
reached by Satlh are naturally completely fanciful; 
those of them who place his birth at the time of the 
bursting of the dam at Marib and his death at 
Muhammad's birth, give him a life of 600 years. It 
should be observed that Abu Hatim al-Sidjistanl (see 
above), whose version is markedly different from the 
others (he does not speak of his monstrosity, puts his 
home in al-Bahrayn, etc.), makes him die in the reign 



of the Himyaritic king Dhu Nuwas and therefore does 
not know of his prophecy to Kisra Anushirwan. 

Bibliography (in addition to works quoted in the 
article): Ibn Hisham, Sim, 9-12; Tabari, i, 911-14, 
981-4; (Noldeke, Geschichle der Perser und Amber, 254- 
7); V-A\ Hi, 312-13 (with variants on the text of al- 
Tabari); Dlnawarl, al-Akhbar al-liwal, 56; Ibn c Abd 
Rabbihi, al-Hkd al-farid' , i, 133-4, 4th ed., i, 94-5; 
Sharh al-Makamal al-Haririyya 2 , i, 216-17 (commen 
tary on the 18th Makdma); DiyarbakrI, TaMkh al 
Khamis, i, 227-8; Mas'udI, Murudj, iii, 364 = 
§ 1249; Kazwfnl, 'AdjaVb al-makhlukdt, ed. 
Wiistenfeld, i, 318-20; Ibn Khallikan, Wafaydt al- 
a'yan, ed. c Abbas, ii, 230-1, tr. de Slane, i, 487-8; 
Damlri, Hayat al-hayawan, 1st ed., i, 46-9, 2nd ed., 
i, 43-4; Freytag, Arabumproverbia, i, 160; Caussin de 
Perceval, Essai sur I'histoire des Arabes anciens , i, 96-7; 
Sprenger, Das Leben und die Lehre des Mohammad, i, 
134-6. A new analysis of the documentation used in 
this article, with some additions, has been made by 
T. Fahd in his La divination arabe, 2 Paris 1987, see 
esp. 44, 66, 83, 101, 165, 186-9, 250. 

_(G. Levi Della Vida-[T. Fahd]) 
SATPANTHIS, adherents of a group in India that 
broke away from the main Nizari Isma c ili da'-wa in the 
Subcontinent in the 10th/16th century. Also called 
Momnas or Imam Shahis, these followers of the Sat- 
panth (the true way) gave their allegiance to Muham- 
mad Shah, the son of Imam Shah, a Nizari Isma c fli 
da'i and son of one of the major pirs of the tradition, 
Hasan Kabfr al-DIn. They continue to preserve their 
own version of the gindns , literary expressions of devo- 
tion and religious teaching common to the da'-wa 
tradition in India, and do not acknowledge the Imams 
of the Nizari line. Their main centres are at Pirana in 
Gudjarat and Burhanpur [q. v. ] in Khandesh. 

Bibliography: W. Ivanow, The sect of Imam Shah 
in Gujarat, in JBBRAS, xii (1936), 19-70; G. 
Khakee, The Dasa Avatara of the Satpanthi Ismdilis and 
Imam Shahis of Indo- Pakistan, diss., Harvard Univer- 
sity 1972, unpubl.; Azim Nanji, The Nizari IsmdHli 
tradition in the Indo-Pakistan Subcontinent, New York 
1978. _ (Azim Nanji) 

SATYA PIR, literally "the true saint", the name 
of a cult which flourished in Bengal and claimed 
both Hindu and Muslim adherents. Islam's entry into 
Bengal gave birth to some important socio-religious 
trends in Hindu society which expressed themselves in 
cults like the Vaishnavite, the Dharma and the Satya 
Pir ones. According to the Shekh Subhodaya, many local 
sadhus, who were mostly Tantric Gurus, embraced 
Islam and adopted devotion to pirs, to whom they at- 
tributed supernatural powers. Cults like Satya Pir, 
Pane Pir, Manik Pir, Ghora Pir and Madari Pir cen- 
tred round pirs and attracted both Hindus and 
Muslims to their fold. 

Tradition ascribes the origin of the Satya Pir cult to 
a Brahmin youth of Mymensingh, Kanka by name, 
who had accepted a Muslim saint as his spiritual 
preceptor. At the behest of his pir he composed during 
the reign of the Bengal Sultan 'Ala 3 al-DIn Husayn 
Shah in ca. 907/1502 an epic poem Vidyasundara kahini 
to pronounce the spiritual glories of Satya Pir. 
Originating in northern and eastern Bengal, the Satya 
Pir cult became popular all over the province. The 
first Bengali poet to compose verses on Satya Pir was 
Shaykh Fayd Allah, whose Satya Pir kavya was compos- 
ed at some time between 952-83/1545-75. Since Fayd 
Allah was an expert in music and had written 
Ragmala, the first book on music in Bengali literature, 
the cult of Satya Pir spread through music and song 
also. Later on, considerable literature appeared on 

Satya Pir in Bengali. Vidyapati, a Hindu poet of the 
18th century, composed his poem Satya Pir ancali, 
which further popularised the cult in Bengal. In the 
later part of the 19th century, the cult lost its earlier 
popularity due to the opposition of Muslim reformist 
movements. The view that Sultan Husayn Shah (898- 
25/1493-1519) was the originator of the Satya Pir cult 
lacks contemporary evidence, though the possibility of 
the sultan encouraging the cult, from mixed motives 
of superstition and policy, cannot be ruled out. 

The followers of the Satya Pir cult make special of- 
ferings on the day of the full moon. They place a 
wooden plank which they consider to be the seat of the 
Satya Pir, and put food and comestibles on it and 
distribute them as tokens of blessing. The Satya Pir 
Bhita (dwelling) stands on the site of the famous Bud- 
dhist monastery at Paharpur in Radjshahi district. Its 
custodian still enjoys rent-free lands, though, barring 
a few areas in Bengal, the Satya Pir cult is non- 

Bibliography: Memoirs of the Archaeological Survey 
of India, no. 55, Delhi 1938, 9, 80; H. Misra, Shekh 
Subhodaya, ed. S.K. Sen, Calcutta 1927; Sukumar 
Sen, Bengala sahitye itihasa, Calcutta 1940, 832; A. 
Karim, Social history of the Muslims in Bengal, Dacca 
1959, 165 ff., 2 Chittagong 1985, 216-18; Bharat- 
chandra, Satya narain vratakatha, Calcutta; Enamul 
Haq, Bange sufi prabhava, Calcutta 1935, 241; idem, 
A history of Sufi-ism in Bengal, Dacca 1975, 277-92; 
D.C. Sen, History of the Bengali language and literature, 
Calcutta 1911, 797; M.R. Tarafdar, Husain Shahi 
Bengal - a socio-political study, Dacca 1965; M.A. 
Rahim, Social and cultural history of Bengal, i, Karachi 
1963, 338-9 etc. (K.A. Nizami) 

SA'UDI-BULAK [see sawdj-bulak] . 
SAUL [see T alut]. 

SAWA (older form Sawadj, cf. the nisba SawadjI, 
found at the side of Sawl), a town of northern Per- 
sia some 125 km/80 miles to the southwest of Tehran 
(lat. 35° 00' N., long. 50° 22' E., altitude 960 
m/3,149 feet). It was formerly on the Kazwln-Kumm 
road used in mediaeval times but now replaced by the 
modern paved roads-system centred on Tehran, and 
on the main caravan and pilgrimage route from south- 
western Persia and lower c Irak to Rayy and 
Khurasan, but this too has been replaced by the 
modern highway from Khuzistan to Arak and Kumm 
and then to Tehran, bypassing Sawa, as does also the 
railway. The town has thus lost importance in modern 
times and the agriculture of the surrounding district 
has been affected by decreasing water supplies and the 
encroachment of salt desert. Sawa itself is situated in 
the northwestern corner of a plain watered by the 
Kara Cay (the mediaeval Gawmaha river) which rises 
in the mountains between Hamadhan and Sawa and 
peters out eastwards into the Great Desert. 
1. History. 

Sawa is not known before the Muslim period. W. 
Tomaschek (Zur historische Topographic von Persien, in 
SB Ak Wien, cii [1883], 154-7) connects its name with 
the Avestan word sava, Pahlavi savaka, "advantage, 
utility" (?). The Persian dictionaries gives "pieces of 
gold" for sawa. According to Tomaschek, Sawa cor- 
responds to the Sevavicina or Sevakina of the Tabulae 

The mediaeval geographers placed Sawa within the 
province of Djibal, but they state that it was attached 
administratively both to Hamadhan and to Rayy at 
various times. It was often linked with the town of 
Aba or Awa [q.v.] to its south on the Gawmaha river. 
The geographers describe Sawa as prosperous because 
of the transit traffic, its camels and camel-drivers 

being famous, and as having a Friday mosque, baths 
and fortifications. The people were strongly SunnI, 
hence often at odds with their neighbours in Awa who 
were fervent Shits (see Hudud al-'-alam, tr. Minorsky, 
133, § 31.22; Le Strange, Lands of the Eastern Caliphate, 
211-13, 228-9; Schwarz, Iran im Mittelalter, 539-42). 
Ibn al-Athir, ed. Beirut, viii, 512, mentions an attack 
of Kurds on the Pilgrimage caravan near Sawa in 
344/955-6. In ca. 420/1029 the lord of Sawa is recored 
as one Kamru al-Daylaml, who allied with the incom- 
ing bands of Ghuzz [q. v. ] to besiege Rayy (ibid., ix, 
382). It suffered badly from sacking by Cingiz Khan's 
Mongols in 61 7/1220, who burnt down a library there 
of what Yakut calls unparalleled richness and which 
also contained, according to al-Kazwml, scientific and 
astronomical instruments. By the next century, how- 
ever, it had revived, and Hamd Allah Mustawff 
describes it as within a fertile agricultural district and 
as flourishing. Its walls were 8,200 ells in cir- 
cumference, having been rebuilt by a local magnate, 
Kh w adja Zahir al-Dfn C A1I b. Malik Sharaf al-DIn 
Sawadjf. The people of Sawa were still Sunnfs of the 
Shafi c I madhhab, but the surrounding villages had 
become largely Shit; the whole district of Sawa pro- 
duced revenue of 25,000 dinars (Yakut, Buldan, ed. 
Beirut, iii, 179-80; al-KazwInf, Athar al-bilad, ed. 
Wustenfeld, 258; Mustawff, Nuzha, 62-3, tr. Le 
Strange, 67-8). 

Among the European travellers, Marco Polo men- 
tions Sawa ("Saba' ') as the town from which the three 
Magi kings set out for Bethlehem and where they are 
buried in a square sepulchre. This Persian-Christian 
legend must be based on a local popular interpretation 
of texts like "Reges Arabum et Saba dona adducent" 
(Psalm lxxii. 10). According to another story given by 
Marco Polo, the three kings are buried respectively at 
Sawa, Awa and Kal'a-yi Atashparastan, which Yule 
located between Sawa and Abhar, while Tomaschek 
identified it with Diz-i Gabran (one stage beyond 
Kum on the road from Kashan). 

Sawa is mentioned by Giosafa Barbara (1474), 
Figueroa (1618), etc. Chardin lamented its sterile soil 
and heat. In 1849 the English consul K.E. Abbot 
counted 300-400 houses in Sawa with 1,000 in- 
habitants; he says that the soil is excellent everywhere 
that it is not mixed with the kawir, but that the salt 
desert is met with at only 6 km/9 miles from the town. 

At the present day the population of the district of 
Sawa is wholly Shit. It consists of Persians and 
Turks. The latter belong to the local confederation of 
Shah-Sewen. which includes the remnants of the tribe 
of Khaladj [q. v. ] . The district of Sawa is frequently 
called Khaladjistan. There are Shahsewen to the 
north-east and to the south of Sawa. The Khaladj live 
more especially to the north of the Kum-Sultanabad 
road (Rahgird, Tadj-Khatun, Djahrud, Tafrish). In 
several of their villages (Kunduriid, Mawdjan, Sift, 
Fowdjird, Kardedjan), various dialects of Turkish are 
spoken, making up the distinct Turkish language 
known as Khaladj, extensively studied by Minorsky 
and Doerfer; see for this, khaladj. 2. Language. In 
recent times Sawa has formed part of various ad- 
ministrative combinations. Sometimes it was gov- 
erned along with the districts to the south (Mahallat, 
Kazzaz), sometimes with Zarand (northeast of Sawa) 
and Kharrakan (vulgo: Karaghan). This last moun- 
tainous district formed an enclave between the prov- 
inces of Kazwin and Hamadhan, It consisted of three 
buluks: Afshar-i Bakishlu, Afshar-i Kutilu and 
Karagoz; the chief town of Kharrakan is situated in 
the latter at the foot of the pass. It is called Awa, and 
must not be confused with the place of the same name 

in the Sawa district. About 1890 Sawa was governed 
by an Austrian officer in the Persian service, von 
Taufenstein. At the beginning of the 20th century, it 
formed a kind of fief of the brigade of Persian 
Cossacks at Tehran. One of the higher officers of this 
military force acted as governor of Sawa and controll- 
ed the Turkish natives who supplied the principal con- 
tingent to the brigade. 

The modern town of Sawa comes within the central 
province (ustan) of Persia and is the chef-lieu of the 
shahrasldn of the same name. The ruins of the 
mediaeval town and its citadel can still be seen. In ca. 
1950 it had a population of ca. 18,000 (Razmara, 
Farhang-i djughrdfiyd-yi Iran-zamin, i, 109), which had 
risen by 1991 to 93,920 (Preliminary results of the 1991 
census, Statistical Centre of Iran, Population 

Among famous men born in Sawa, Yakut mentions 
Abu Tahir <Abd ai.Rahman b. Ahmad, one of the 
principal Shafit imams (d. 484/1091). A colleague of 
the great vizier of the Mongols, Rashld al-DIn [q.v.], 
was Sa<d al-DIn Sawadji, executed in 711/1312 after 
his fall from power. Mustawff mentions the tomb of 
Shaykh c Uthman Sawadji near the town. On the poet 
Salman-i Sawadji (700-78/1300-76), see E.G. 
Browne, LHP, iii, 260-71 etc., and the article s.v. 

These include: (i) The barrage on the Kara-cay 
(about 12 miles south-south-west of the town), said to 
owe its origin to Shams al-DIn Djuwaynl [q.v.], vizier 
of several Il-Khanid rulers of the 7th/ 13th century (see 
Mustawff, Nuzha, 221). The barrage is said to have 
been restored under the Safawids; it is known as band-i 
Shah '■Abbas. It occupies the passage between two hills 
and is about 65 feet high, 100 long and 45 thick. 
Beside it on the left bank, the road rises in a kind of 
spiral; caravans were thus able to ascend the dam, 
which was used as a bridge, and descend on the west 
side by a gradual slope on the right bank. 

(ii) The fortress of Kiz-kal c a on a rock in the centre 
of an amphitheatre of hills not far from the dam. 

(iii) The mosques and minarets of the town itself. 
The Friday Mosque on the south side of the town, 
rebuilt in the Safawid period by Shah Tahmasp I, 
may well occupy the site, so Minorsky thought, of the 
Friday Mosque mentioned by al-MukaddasI, 392 n. a 

t froi 

t"); i 


cient structure, and Buyid period inscriptions (so far 
unpublished) have recently been uncovered there. 
Adjacent to it, but free standing, is a brick minaret, 
first described by Dieulafoy, with an inscription dated 
504/1110-11 in the names of the Saldjuk sultan 
Muhammad b. Malik Shah and the c Abbasid caliph 
al-Mustazhir [q.vv.\. In the town centre is the 
Masdjid-i Maydan with an early Safawid inscription 
and with its minaret naming as builder a local 
DaylamI amir, Abu Dulaf Surkhab b. c Imad al-Dawla, 
and the date 503/1109-10. 

Bibliography: W. Tomascheck, Zur historischen 
Topographic von Persien, i, in SB Ak Wien, cii (1883), 
154-7; Le Strange, Lands of the Eastern Caliphate, 
210 ff., 228-9; Schwarz, Iran im Mittelalter, 539-42 
(these authors giving references to the Arabic 
geographers); Hudud al-'-alam, tr. Minorsky, 133; 
Mustawfl, Nuzha, 62-3, tr. 67-8; K.E. Abbott, 
Geographical notes ..., in JRGS (1855), 4-10; H. 
Binder, Au Kurdistan, Paris 1887, 380 (photograph 
of the ab-anbar at Sawa); J. Dieulafoy, La Perse, 
Paris 1887, photographs of the barrage and 
religious buildings within the town, 165-173; Sir A. 
Houtum-Schindler, Eastern Persian Irak, London 
1903, 129-30; Yule-Cordier, The Book of Ser Marco 


Polo 3 , London 1903, i, 78-81; F. Sarre, Denkmdler 
persischerBaukunst, Berlin 1910, ii, 112-23; E. Herz- 
feld, Khorasan. Denkmalsgeographische Studien zur 
Kulturgeschichte des Islams in Iran, in 1st., xi (1921), 
170; Pope, Survey of Persian art, index; A. Godard, 
Les anciennes mosquees de I'lran, in Arts asiatiques, iii 
(1956), 48-63, 83-8; G.C. Miles, Inscriptions on the 
minaret of Saveh, Iran, in Studies in Islamic art and ar- 
chitecture in honour of Professor K.A.C. Creswell, Cairo 
1965, 163-78; Sylvia A. Matheson, Persia, an ar- 
chaeological guide 2 , London 1976, 190. 

(V. Minorsky-[C.E. Bosworth]) 
3. The town's role in Muslim legend. 
Sawa plays an important part in the legends of 
Muhammad. According to a frequently quoted 
tradition (for details see A. Sprenger, Das Leben und die 
Lehre des Mohammad, i, 134 ff., and Th. Noldeke, 
Geschichte der Perser und Araber, 253 ff.), a lake (buhayra) 
in the neighbourhood of Sawa sank into the ground in 
the night in which the Prophet was born. The site was 
still pointed out to al-Kazwinl in the 7th/13th century. 
As the tradition quoted shows a rather accurate 
knowledge of Iranian matters, we may safely seek an 
allusion to a definite Iranian conception in this single 
feature of the story. Now in Zoroastrian eschatology 
the lake Kansava (Kasaoya-) plays an important part; 
in the later Avesta it is located in Eastern Iran and is 
said to correspond to Lake Hamun in Sidjistan. In it 
is preserved the seed of Zarathushtra, from which in 
the end will arise the saviour Saoshyant. When we 
find the legend of the drying-up of a lake in Iran con- 
nected with the birth of Muhammad, we may inter- 
pret it as an allusion to this mythical lake. The legend 
symbolises the destruction of the hope of a Zoro- 
astrian saviour, just as the earthquake in the royal 
palace at Ctesiphon recorded in the same tradition 
symbolises the end of the Iranian empire and the ex- 
tinction of the sacred fire the end of Zoroastrian 
culture. _ (H.H. Schaeder) 

SAW AD, a name used in early Islamic times 
for 'Irak [q.v.]. While the name Irak has been 
proved to be a Pahlavi loanword (from Erag, "low 
land, south land", occurring in the Turfan fragments, 
with assimilation to the semantically connected root 
c rk; cf. A. Siddiqi, Studien uber die per sischen Fremdwbrter 
im klass. Arab., Gottingen 1919, 69; H.H. Schaeder, 
in 1st., xiv, 8-9; J.J. Hess, in ZS, ii, 219-23) sawed 
"black land" is the oldest Arabic name for the alluvial 
land on the Euphrat« 


o the e : 

: betw, 

d the 

dazzling white Arabian desert (Yakut, jWu c . 
174, 14 sqq.). The name has undergone a threefold 
development of application. (1) It is identified with 
the political notion of c Irak and thus corresponds to 
the Sasanid province of Suristan (Dil-i 
With this meaning, the historians of the Arab con- 
quests use the name Sawad for c Irak (see e.g. al- 
Baladhurl, Futuh, 241, 1; because the Muhadjirun ac- 
quired such extensive properties there, it became 
known as "the Garden of Kuraysh"), and, especially, 
the compilers of monographs on taxation or political 
handbooks (cf. Abu Yusuf, Yahya b. Adam, Kudama 
and al-Mawardl; also Ibn Khaldun). The reason for 
this is that in the cadastral and revenue regulations of 
c Umar I, the name Sawad was used officially. (2) It is 
used as the name of the cultivated area within a 
district, e.g. Sawad al- c Irak, Sawad Khuzistan, Sawad al- 
Urdunn. (3) Before the name of a town, it means the 
systematically irrigated and intensively cultivated 
fields in its vicinity, e.g. the Sawad of al-Basra, Kufa, 
Wasit, Baghdad, Tustar, Bukhara, etc. 

Bibliography: The fundamental work is H. 

Wagner, Die Uberschdtzung der Anbauflache Babyloniens 
und ihr Ursprung, in Nachrichten v.d. Kgl. G. W. Gott. , 
Phil. -hist. Kl. (1902), 224-98. On the philological 
point, see Lane, i, 4, 1462b. On the technical ques- 
tion of taxation, A. von Kremer, Uber das Budget der 
Einnahmen unter der Regierung des Harun al-rasid, in 
Verh. d. VII. Internal. Orient. Kongr. , ii, 1888; M. van 
Berchem, La propriete territorial et I'impot fonder sous 
les premiers califes (1886); D.C. Dennett, Conversion 
and the poll tax in early Islam, Cambridge, Mass. 
1950, ch. II; F. Lokkegaard, Islamic taxation in the 
classic period, Copenhagen 1950, index; M.G. 
Morony, Iraq after the Muslim conquest, Princeton 
1984, index. (H.H. Schaeder*) 

SAWAFI [see safi]. 

SAWAKIN, the Arabic form, in English Suakin; 
Bedawytt: u-Sok, an island and port on the west 
coast of the Red Sea, at 19° 07' N. and 37° 20' E. 
in the present Republic of the Sudan. 

A narrow inlet from the sea, about 3 km/2 miles 
long, widens on the inside to form a natural harbour 
11-15 m/35-50 feet deep which can easily accom- 
modate about 20 traditional coastal vessels; growing 
coral reefs make it dangerous to navigate at night. In 
this harbour, a coral island (Djazlrat Sawakin), 360- 
450 m /l, 200-1, 500 feet in diagonal, rises ca. 2 m/6K 2 
feet above high water mark. Another two small and 
one larger island (Djazlrat al-Shaykh c Abd Allah al- 
Djabartl, marked on British maps variously as 
Condenser-, Hospital-, or Quarantine Island) are 
connected to the mainland today by silt and salt-pans 
and no longer recognised as islands. Djazlrat Sawakin 
is the site of the old town, a typical Arab-Islamic In- 
dian Ocean merchant settlement. Its houses, built 
mainly of coral, are up to four storeys high; most have 
tumbled down today. A causeway built in 1878 links 
the island to the mainland (al-Kayf; Bedawytt: u- 
Gef). There, the dwellings are mostly Bedja [q. v. ] huts 
built of reed, palm mats, driftwood, and more recent- 
ly, scrap metal or sackcloth. Much of al-Kayf is sur- 
rounded by the remnants of a fortified wall erected 
1887-8. Outside the wall a number of suburbs have 

al-Fula and the salty wells at al-Shata. The mainland 
settlement has been more populous than the island 
town since at least the early 19th century; in 1814, the 
number of its inhabitants was estimated at 5,000, 
compared to 3,000 on the island. The population grew 
only slightly during the century and fell to ca. 5,000 
during the Mahdist wars. After a brief recovery 
(1904: 10,500) it dwindled to around 4,000 during 
1937-67, but rose again in the 1970s. The local census 
of 1979 gave a figure of 13,714 inhabitants (all on the 
mainland). Sudanese Bedja and refugees (often from 
Eritrea) each constituted about 40%. 

Popular etymology has inspired several legends 
concerning the origin of Sawakin. The best-known 
story tells of seven virgin slave girls sent from Ethiopia 
to a northern king. Upon arrival they were found to 
be pregnant and explained that during an overnight 
stay on the island of Sawakin they had been visited by 
"seven djinn" (sab c a djinn), and that "the Djinn had 
done it" (sawwa djinn [sic]), for it was "the djinn s 
dwelling" (sakan al-djinn). The king sent them back to 
the island where their children were born, and from 
them descended the inhabitants of "Sawadjin" > 
Sawakin. The lexicographer al-FIruzabadT lists the 
island in his Kamus s.v. s-k-n. Roper argued that the 
name is derived from suk "market" (Bedawytt u-suk, 
construct form i-sukib, which may be heard as i-sookim; 
Correspondence, in Sudan Notes and Records, xxii [1939], 

The port, one of the few sheltered anchorages with 
a fairly reliable supply of water on an otherwise in- 
hospitable coast, has probably been used since Anti- 
quity. Attempts to identify it with place names col- 
lected by Greek and Roman geographers from the 
heterogeneous information brought back by mer- 
chants have, however, remained inconclusive. Most 
likely candidates are AiuY|v EuxpfeXioov "port of good 
tidings", 'A<m(? "round shield", and Saxo XiuV|v (all 
from Ptolemy; the latter is found on Tabula XV, 
Codex Urbinas Graecus 82, fols. 87 (86) v -88 (87/), and 
Sace/Sacae/Suche of Pliny. The name Sawakin is first 
mentioned in the 4th/ 10th century as a minor port in 
the country of the Bedja with trade connections to 
Djudda and Nubia (al-Hamdam, 41, 133; Ibn 
Hawkal, 55; Yakut, iii, 182; Ibn Sulaym ai-Uswanl, 
Akhbdr al-Nuba, in al-MakrizI, K. al-MawaHz, ed. G. 
Wiet, Cairo 1922, iii, 257, 272). The export of grain 
and animal products from its hinterland to the Hidjaz 
was probably the earliest and most stable source of 
Sawakin's income, but its natural harbour also gave 
it a certain significance as an anchorage on the trade 
route between Egypt and the Yemen, even while 
c Aydhab [q. v. ] was still the largest port on the 
Sudanese coast. This significance increased in the 
6th/ 12th century after the decline of Badi c , and 
Sawakin subsequently became involved in the strug- 
gles for control of the Red Sea trade between Dahlak 
[q.v.], Egypt, and the Meccan Sharlfs. The Fatimids 
established a special fleet to protect the ships of the 
Karimi [q.v.] merchants threatened by pirates while 
travelling between c Aydhab and Sawakin. On occa- 
sion, Sawakin and Dahlak went to war over control of 
the islands lying between them. Sawakin's local rulers 
(Muslims since at least the 7th/13th century) collected 
customs from the ships calling at the port; they had, 
however, to acknowledge the suzerainty of the rulers 
of Egypt and to pay them tribute. In case of insubor- 
dination they faced punitive expeditions. 

By the early 8th/14th century the Sharlfs of Mecca 
[q.v.] gained control over Sawakin (Ibn Battuta, ii, 
161-2, tr. Gibb, ii, 362-3), perhaps driven by a desire 
to secure a trade route that reduced their dependence 
on grain imports from Egypt. They were only able to 
establish a foothold on the African coast by co- 
operating with the Bedja, whose rulers (the Hadarib) 
maintained direct control of the more important port 
of c Aydhab. Both sides had to acknowledge a measure 
of Egyptian sovereignty. 

Following the decline of c Aydhab, Sawakin became 
the largest northeast African port north of Masawwa c 
[q.v.], serving as an entrepot for the upper Nile basin 
and as the major point of embarkation for pilgrims 
from the Sudanic belt. Through the commercial net- 
works of the Banya (Hindu and Jain merchants from 
India) Sawakin partook in the lucrative (but fluc- 
tuating) Indian Ocean trade. Towards the end of the 
8th/ 14th century the Hadarib moved their capital to 
Sawakin. Egyptian pressure continued; Sawakin was 
sacked in 843/1439-40 and had to submit to the 
authority of the Mamluks (who used it as a place of 
banishment) until the end of their rule. 

The 10th/16th century saw two major changes in 
the geopolitical situation. The Ottomans became the 
leading military power in the Red Sea after their con- 
quest of Egypt and the defeat of the Portuguese, and 
they succeeded the Mamluks as overlords of Sawakin. 
Their first landing there occurred around 1520, and 
in 962-4/1555-7 Ozdemir Pasha [q.v.] established the 
Eyalet of Habesh [q.v.], with Sawakin as its ad- 
ministrative headquarters. The residence of the 
beylerbeyi was transferred several times between 

Sawakin, MasawwaS and Djudda until 1113/1701, 
when the province was finally united with the sandjak 
of Djudda and the post of shaykh al-Haram of Mecca, 
with Djudda as seat of the administration. The ka Hm- 
makdm of the sandjak of Sawakin was appointed by way 
of annual tax-farming. 

Secondly, the Fundj sultanate of Sinnar [?.»».], the 
new political centre on the Nile, tried to gain control 
of the port, which led to fighting with the Ottomans. 
Neither of these outside forces was able to maintain 
complete authority over Sawakin, however; they had 
to come to terms with the Bedja amir who resided on 
the Kay f (the families monopolising this office became 
known as "Artega"). In theory, the amir shared the 
revenues of the port (customs and a poll-tax for 
pilgrims) with the Ottomans; in practice, he only paid 
if he felt they were strong enough to pose a threat. 

The diversion of the trade route from India to 
Europe around Africa caused a depression of 
Sawakin's international trade in the 17th-18th cen- 
turies. But its function as a trans-shipment centre for 
African trade with the Arabian peninsula secured it 
continued importance, and when Ottoman power in 
the Red Sea declined at the end of the 18th century, 
the Sharif of Mecca assumed control of Sawakin. 
During the first decade of the 19th century, the power 
struggles between the Sharif, the Wahhabls, the Ot- 
tomans and the Egyptians gave greater independence 
to the local leaders in Sawakin. But when Muhammad 
C A1I [q.v.] defeated the Wahhabls in 1813, Sawakin 
became subject to Egyptian domination (though not 
sovereignty) even before the conquest of the Sudan. 
In 1830 the Ottoman Sultan officially recognised this 
situation, but when Egypt had to evacuate most of her 
acquisitions in 1841, the Red Sea territories were 
nominally returned to the Ottomans. They were 
unable to exert effective control, however, leaving 
Sawakin and Masawwa' practically independent. 
Late in 1846, the Ottomans leased both ports to 
Muhammad C A1T for the duration of his lifetime 
against payment of an annual tribute of 5,000 purses 
(the amount suggests that Egypt had strategic rather 
than economic interests). In 1849, they reverted to the 
Ottomans, though an actual Ottoman presence was 
not re-established until 1851. Egypt continued her ef- 
forts to gain control of Sawakin, which many regarded 
as an easier outlet for the increasing Sudan trade than 
the Nile route. In spite of protests by local traders, 
who feared Egyptian competition, Sawakin and 
Masawwa' were transferred to Egyptian administra- 
tion in Dhu '1-Hidjdja 1283/May 1865 (for details, see 
G. Douin, Histoire du regne du Khedive Ismail, iii/ 1 , 
Cairo 1936, 268-77, 305-6). 

Sawakin's importance grew with the increase in 
Red Sea trade brought about by the introduction of 
steam shipping (since 1829), the colonial exploitation 
of the interior as a source of gum, ivory, and slaves, 
and the marketing of European industrial goods. 
After the opening of the Suez Canal (1869), European 
merchants were able to outdo their Indian and Arab 
competitors; the latter were increasingly left with 
smuggling and the slave trade (formally illegal since 
1855). Nevertheless, the export of ghee and grain to 
the Hidjaz remained the largest and most stable part 
of Sawakin's trade. 

Political stability and the economic opportunities 
attracted a cosmopolitan population, and the building 
boom of the 1870s effaced much of the earlier ar- 
chitecture. The over-ambitious modernisation pro- 
jects devised by the Egyptian administrators largely 
failed, however, and the exaction of forced labour, 
heavy taxation, and increasing "foreign" control did 


not help the government's popularity. A significant 
part of the local population left the town during the 
Mahdiyya [q.v.], and from 1884-91 Sawakin, the only 
Sudanese town never conquered by the Mahdists, was 
the chief base for British military operations in the 
eastern Sudan. Trade was severely disrupted, though 
it never stopped. The important route to Berber was 
reopened in 1891 , and for a brief time Sawakin flour- 
ished again as an import centre for the new Con- 
dominium authorities. 

Natural deficiencies made it unfit for modern ship- 
ping, however, and after 1905 it was replaced by the 
newly-constructed Port Sudan which, apart from 
having more room for expansion, gave the British 
greater control over maritime commerce by limiting 
the role of Sawakin's indigenous urban population 
and of the Egyptians. By 1910, the administrative and 
economic centre of gravity had shifted to the new 
port, and after most merchants transferred there 
(1923-5), Sawakin fell into disrepair. In the mid- 
19405, over 80% of its houses had collapsed. The 
customs were finally closed in 1946, and Sawakin was 
left with only the pilgrimage traffic which was official- 
ly restricted to the old port during 1908-52 and 
lingered on until 1973. 

The relocation of the Sudan's official dhow harbour 
to Sawakin in 1968-72 gave the town a new function 
as a centre of boatbuilding, dhow trading, and 
fishing. Its social composition was radically altered by 
the influx of drought and war refugees during the 
1970s; the island town remains deserted, and Sawakin 
now consists of a collection of villages on the 
mainland. A new impetus has been given by the large 
container port opened in 1991 to relieve pressure on 
Port Sudan. 

Bibliography (if no page numbers are given, see 
index s.v.): Mustafa Muhammad Mus'ad, al- 
Maktaba al-Suddniyya al- c Arabiyya, Cairo 1972 (com- 
prehensive collection of mediaeval Arabic texts 
relative to the Sudan); J.F.E. Bloss, The story of 
Suakin, in Sudan Notes and Records, xix (1936), 271- 
300, xx (1937), 247-80 (chronological survey, chief- 
ly of travellers' accounts in European languages); 
O.G.S. Crawford, Ethiopian itineraries circa 1400- 
1524, Cambridge 1958; J. de Castro, Roteiro do Mar 
Roxo, Lisbon 1991; J. Lobo, Itinerario, Oporto 1971, 
Eng. tr. D.M. Lockhart, London 1984; The Red Sea 
and adjacent countries at the close of the seventeenth century, 
ed. W. Foster, London 1949, 107, 154, 176; J. L. 
Burckhardt, Travels in Nubia, London 1819, 431-58, 
510; W. Munzinger, Ostafrikanische Studien, Schaff- 
hausen 1864, 109-12; R. von Neimans, Das rothe 
Meer . . . imjahre 1857 in handelspolitischer Beziehung, in 
ZDMG, xii (1858), 391-438; T. von Heuglin, Reise 
in Nordost- Africa . . . imjahre 1857, in Mittheilungen aus 
Justus Perthes' Geographischer Anstalt, i (1860), 336-8; 
Shawkl al-Djamal, al-WathaHk al-la\tkhiyya li-siyasat 
Misr fi 'l-Bahr al-Ahmar (1863-1879), Cairo 1959, 
29-178; E. Sartorius, Three months in the Sudan, Lon- 
don 1885; N. Shukayr, TaKkh al-Suaan al-kadim 
wa'l-hadith wa-djughrdfiyatuhu, Cairo 1903; The 
Anglo-Egyptian Sudan, ed. Count Gleichen, London 
1905; J.R. Crowfoot, Some RedSea ports in the Anglo- 
Egyptian Sudan, in Geogr. Jnal., xxxvii (1911), 530 
(on Antiquity); Naval Intelligence Division, Western 
Arabia and the Red Sea, London 1946; O.G.S. 
Crawford, The Fung kingdom of Sennar, Gloucester 
1951, 118-26, 175-9; D.H. Matthews, The Red Sea 
style, in Rush, i (1953), 60-86; idem, Suakin postscript, 
in Rush, iii (1955), 99-111; Yusuf Fadl Hasan, The 
Arabs and the Sudan, Edinburgh 1967; D. Roden, The 
twentieth century decline of Suakin, in Sudan Notes and 

Records, li (1970), 1-22; A. Paul, A history of the Beja 
tribes of the Sudan, London 1971; R.S. O'Fahey and 
J.L. Spaulding, Kingdoms of the Sudan; C. Orhonlu, 
Habes eyaleti, Istanbul 1974; J. -P. Greenlaw, The cor- 
al buildings of Suakin, Stocksfield 1976; G.H. 
Talhami, Suakin and Massawa under Egyptian rule, 
1865-1885, Washington 1979; Muhammad Salih 
Dirar, Ta'rlkh Sawdkin wa'l-Bahr al-Ahmar, Khar- 
toum 1981; M. Daly, Empire on the Nile, Cambridge 
1986; E.G. Kentley, Suakin and its fishermen, Ph.D. 
diss., University of Hull 1988, unpubl.; K.J. 
Perkins, Port Sudan, Boulder 1993. 

(A. Hofheinz) 
SAWDA bt. ZAM'A b. Kayyis b. c Abd Shams, 
Muhammad's second wife, was one of the early 
converts to Islam. She was of the noble tribe of 
Kuraysh [q. v. ] on her father's side, and her mother al- 
Shamus bt. Kayyis b. c Amr of the Banu c Adi, was 
from the Medinan Ansar [q.v.], supporters of the 
Prophet. She accompanied her first husband al- 
Sakran b. c Amr to Abyssiniya, along with her brother 
and his wife, with the second party of Muslims who 
emigrated there. The couple returned to Mecca before 
the hidjra [q.v.], and al-Sakran, who had become a 
Christian in Abyssinia, died there. 

Sawda's marriage to Muhammad was arranged by 
Khawla bt. Hakim [q.o.], who wished to console him 
for the loss of Khadldja [q.v.], and took place about a 
month after the latter's death, in Ramadan 10/631. 
There is some disagreement among Muslim scholars 
whether the Prophet first married Sawda, the mature 
widow, or the virgin bride 'A'isha [q.v.]. According to 
one tradition, he chose Sawdabecause of her devotion 
to his mission rather than 'A'isha, daughter of his 
closest friend. In any case, Sawda lived alone with 
Muhammad for some time since 'A'isha was too 
young for her marriage to be consummated and re- 
mained with her parents. 

In the first year of the hidjra, Sawda and Muham- 
mad's daughters were brought to join him in Medina; 
her dwelling was among the first to be built adjacent 
to the mosque area. Despite her access to the Prophet 
and proximity to the centre of affairs, only one or two 
traditions are related from or about her compared to 
the numerous 'hadiths ascribed to 'A^isha. On two oc- 
casions when Sawda asserted herself, Muhammad 

The most noteworthy event in Sawda's life with 
Muhammad was her success in dissuading him from 
resulted in the revelation of 
in feareth ill-treatment from 
in ... it is no sin for them 
twain if they make terms of peace between them- 
selves." Muhammad favoured c A>isha and (in 8/629) 
apparently decided to divorce Sawda, but she stopped 
him in the street and implored him to take her back. 
She offered to yield her day to '■A'isha, as "she was 
old, and cared not for men; her only desire was to rise 
on the Day of Judgement as one of his wives". The 
Prophet consented, and after that the verse was re- 
vealed to him. 

Sawda gained a reputation for being heavy and 
ungainly because during a pilgrimage, Muhammad 
allowed her to reach Mina for the morning prayer 
before the crowd's arrival, to avoid being jostled, a 
privilege he denied 'A'isha. Because she was an 
unusually tall woman, 'Umar recognised her from 
afar at night and convinced the Prophet to seclude his 
wives. Sawda is sometimes confused with Zaynab bt. 
Djahsh [q.v.] as the "longest-handed" among the 
Prophet's wives, i.e. the most charitable, who would 
be the first to join him in Heaven. She was generous 

divorcing her, because 
sura IV, 127: "Ifawoi 
her husband, or desen 

and humorous. 'A'isha used to say: "There is no 
woman in whose skin I would rather be than Sawda's, 
except that she is somewhat envious". Sawda as well 
as Zaynab bt. Djahsh did not take part in the last 
pilgrimage, implying that God and his Prophet pre- 
ferred them to remain in their home. 

After Muhammad's death, Sawda received a gift of 
money from c Umar, which she ordered to be dispers- 
ed like her share of the spoils of Khaybar. She died in 
Medina, in Shawwal 54/September-October 674, 
during the caliphate of Mu c awiya, who bought her 
house in the Mosque precincts, together with that of 
Safiyya, for 180,000 dirhams. 

Bibliography: Ibn Hisham, 214, 242, 459, 787, 
1001; Ibn Sa c d, viii, 35-39; Ibn Hanbal, Musnad, 
vi, 223, 271; Tabari, i, 1767, 1769, hi, 2437-40; 
Aghanl 1 , iv, 32; Ibn Hibban, K. al-Thikdt, Hay- 
darabad 1973-82, iii, 183; Ibn al-Kaysaram, al- 
Diam c bayna kitdbay AbiNasr al-Kaldbddhi wa-Abi Bakr 
al-Isbahdni fi ridjdl al-Bukhdri wa-Muslim, Haydar- 
abad 1905, ii, no. 2371; Dhahabi, Siyar aHam 
al-nubald\ Beirut 1981-8, ii, no. 40; Ibn Hadjar 
al- c Askalam, Isdba, Cairo 1970-2, vii, nos. 11357, 
11887; Caetani, Annali dell' Islam, i, 378-9; N. 
Abbott, Aishah, the beloved of Mohammed, Chicago 
1942, 3-8, 11-12, 24, 26, 41-42, 44-45, 94, 197; 
Sawda, Cairo, Dar al-Ma c arif 1983. 

(V. Vacca-[Ruth Roded]) 
SAWDA, Mirza Muhammad Rafi c (1125- 
1195/1713-1781), a highly esteemed Urdu poet, was 
born in Dihli. His father came from a military family 
of Kabul, and he settled in Dihli, where he became a 
wealthy merchant. The future poet was a spendthrift 
in his youth, and after his father's death he soon 
disposed of his inheritance by riotous living. After a 
spell of soldiering, he turned to a poetical career, 
adopting the takhallus of Sawda (Ar. "melancholy, 
madness"), an apt name in an age when poets con- 
centrated on ghazal. Perhaps it was also a pun on Per- 
sian sawda ("trade"), in allusion to his father's oc- 
cupation (Saksena, op. cil., 61). His courtly manners, 
due to his upbrining, enabled him to find patrons easi- 
ly, and his undoubted poetic talents soon won him 
recognition. As was the usual custom, he_concentrated 
largely on Persian: but the poet Khan-i Arzu persuad- 
ed him to change to Urdu "a momentous decision 
which had a far-reaching effect on his poetic career 
and the development of Urdu poetry" (Sadiq, 83). 
Saksena (op. cit. , 66) describes him as the 
"Shakespeare of Urdu"). And few will deny that he 
is a pioneer of Urdu poetry, and probably one of the 
six greatest poets of the 18th and 19th centuries. 

Though he composed many ghazal poems — 
especially in his early years — he is not generally 
regarded as a master in the genre. Perhaps it would 
be more accurate to say that he does not project the 
accepted portrait of the ghazal-gu as a lover let down 
by a cruel beloved. Nor, at the other end of the ghazal 
spectrum, can he be regarded as a mystic poet. He 
was a Shr c i and exceedingly proud of it. This emerges 
in his poetry in various forms, but it is not the key to 
his mastery as a poet. This lies in his command of 
many different genres, his virtuosity in the tech- 
niques, his positive approach to his metier and his ap- 
parent pride and joy in it. 

The political situation in Dihli made life there in- 
creasingly less congenial, and an exodus of poets 
gathered momentum. In 1754 Sawda moved to Far- 
rukhabad: then in (?) 1771 to Faydabad (Fayzabad), 
where the Nawab Shudja c al-Dawla became his 
patron. The succeeding Nawab Asaf al-Dawla found- 
ed a new capital at Lucknow in 1774, and took the 

poet there with him, granting him a substantial salary 
for the rest of his life. 

Though undoubtedly an architect of Urdu poetry, 
he retained many features of "Persian decadence", as 
later critics called it, but not, however, the passion for 
ihdm (double entendre). His vocabulary is rich, but 
predominantly Persian-Arabic. He revelled in dif- 
ficult rhymes and metres. His poetry is full of energy, 
and at time vitriolic. Turning to the forms for which 
he is famous, he is recognised as a— possibly the— 
leading kasida-gu (writer of odes) [seeMApiH, madh. 6. 
in Urdu]. These include religious ones in praise of the 
Prophet and the Shr c I Imams, and secular ones in 
praise of nobles and rulers including Nawab Ghazi al- 
Dln Khan on his birthday, and Nawabs Shudja c al- 
Dawla and Asaf al-Dawla. There are also numerous 
poems of satire (hadjo/hadjw), which ranged from the 
amusing to the vulgar, and are often exaggerated. He 
also composed a number of mathnawis and marthiyas. 
His poems in murabba c a and mukhammas and various 
other forms should be mentioned. He is also credited 
with introducing into Urdu verse what is termed 
tamthiliyya shdHri ("gnomic verse"), in which the 
thought expressed in the first hemistich of a verse is 
followed by an illustrative metaphor or simile in the 
second; this could be used in many verse forms. 

Bibliography: Among a number of published 
collections of his poetry, Kulliyydt Sawda by 
Nawal Kishor, Lucknow 1936, can be recommend- 
ed. Of critical studies of the poet, Khalik An- 
djuman, Mirza Muhammad Raft'' Sawda, Aligarh 
1966, deserves mention. A useful selection of his 
poetry is found in Kadrat C A1T Shawk, Tabakdt al- 
s_hu<ara>, ed. Nadhlr Ahmad Faruk, Lahore 1968, 
119-71. Unfortunately, the earlier pages, including 
the biographical notes are missing from the original 
ms. of this section. Accounts of the poet are 
found in two general English sources for Urdu 
literature, namely Muhammad Sadiq, A history of 
Urdu literature, London 1964, 82-93, with many 
quotations in Urdu script, often with English 
translations: also Ram Babu Saksena, A history of 
Urdu literature, Aligarh 1927, 60-80, which contains 
no quotations. 0-A. Haywood) 

al-SAWDA 5 , more precisely, al-Khariba al- 
Sawda' "the black ruin", the ancient Nashshan, an 
important archaeological site in the Djawf of 
Yemen, 102 km/63 miles to the north-east of San c a\ 
still inhabited today, and adjoining the north-east of 
the modern settlement of al-Maslub, the main centre 
of the Banu Nawf. It lies on the left bank of the Wadf 
'1-Kharid (locally called al-Buhayra), and the tell 
dominates the surrounding plain by a dozen metres. 
The ancient enceinte, ca. 1,200 m/3,940 feet long, 
encloses a rectangle 330 m/1 ,080 feet by 280 m/920 
feet. It comprises a thick massif of unfired bricks on 
to which abut two walls faced with stone, one very 
carefully constructed facing the outside, and the other 
less so facing the interior; of these, there remains only 
a few stone foundations and a pile of sun-dried bricks. 
As in other sites of the Djawf, the curtain wall was 
reinforced at regular intervals with square-plan 
towers; several of these are still standing in the 
western part, the best preserved. There were four 
gates, all now in a very ruinous state, which can be 
discerned in the middle of each of the sides. Inside the 
town, the main visible traces are large footings of 
regularly-cut stone and pillars or architraves belong- 
ing to the gateways. 

At 720 m/2,360 feet to the east of the town, a fine 
sanctuary, dedicated to the local god c Athtar dh u 
Risaf (Htr d-Rsf) has been almost wholly preserved by 

the s, 

rtially e 

ited by the French Archaeological Mi 
Yemen Arab Republic in the winter of 1988-9 
(Breton, 1992), and is notable for the quality of its ar- 
chitecture and its decoration, incised on the supports 
of the lintel of the entrance porch, as well as the 
monolithic pillars supporting the stone covering. The 
main figures of this decor are young women on a 
podium (which_ the people in the Djawf call 
"Daughters of c Ad", referring to the mythical tribe of 
that name [see c ad]), ibexes in movement, ostriches, 
snakes interwined in pairs, bulls' heads and geometric 
motifs. The very ancient inscriptions carved on the 
pillars of the entrance porch (SW/BA/I/1-4) and 
carbon-5 analysis of the wood found in the mortar of 
the enclosing wall allow us to date the oldest parts of 
the temple to the 8th century B.C. 

The irrigated perimeter of the wall round the tell is 
now hardly discernible. The traces of a tamarisk 
thicket (athf) which struck the discoverer of the site, 
Hayyim Habshush (ed. and summary tr. Goitein, Ar. 
text 52/7-8; Moscati-Steindler, 1976, 86) nevertheless 
show that a good site for agriculture and a water sup- 
ply existed there. Irrigation seems to have been prac- 
ticed for many centuries, judging by the thickness of 
the alluvium cut through by the wadis, which exceeds 

The site and the temple of c Athtar have yielded 87 
inscriptions, in the two epigraphic South Arabian 
languages Madhabi and Sabaean (see on these, 
Robin, 1992, 31 ff.; for a complete list of these, see 
Avanzini, 1995). 

Al-Sawda 3 is the ancient Ns 2 n, shown by four in- 
scriptions at the site containing this name (al-Sawda 3 
52 = CIH 440/4; al-Sawda 3 75 = RES 2902 = M 
126/1; SW-BA/I/5 and 6); the vocalisation of the 
name was probably Nashshan, if one relies on the sole 
mention of this toponym in Arabic, a poem of the 12th 
century A.D. (Madelung, 1992, 37). This last was the 
capital of a small independent kingdom, equally called 
Nashshan, in the 8th century B.C. and at the begin- 
ning of the 7th one. Only one person with the title of 
king is known: Sumhuyafa c Yasran, son of Luba 3 an, 
king of Nashshan {S'mhyf Ys'rn bn Lb* mlk Ns 2 n, in 
SW/BA/I/5 and 6; see also RES 3945/14-17 and al- 
Sawda 3 4, where the same person is mentioned with- 
out a title). The kingdom's territory included the town 
of Nashshan. that of Nashk um {Ns 2 lf, modern al- 
Bayda 3 ) and a district called Ayk um (>f"), with 
numerous settlements. At the beginning of the 7th 
century B.C., the kingdom of Saba 5 , which already 
dominated the basin of the Adhanat Cdnl, modern 
wadl Dhana) river and the northern sector of the 
Djawf, undertook to impose its protection over Nash- 
shan, probably fearing that the latter would unify the 
Djawf and become a threat to Sabaean supremacy in 
South Arabia. Led by a mukarrib (a sovereign superior 
to the kings, whose name means "unifier") Karib 3 il 
Watar, son of Dhamar c ali, Saba' launched two suc- 
cessful campaigns against Nashshan, which emerged 
from the struggle seriously weakened, losing territory 
to Saba 3 (Nashk um and Ayk um ) and suffering under 
severe conditions (destruction of the royal palace 
called c Afraw ( c /™) and the enceinte of the capital, in- 
stallation of a Sabaean garrison and the building of 
the temple of Almakah (Hmkh), the great god of the 
Sabaeans, in the centre of the town (see on this war, 
Robin, forthcoming). Its independence was thus com- 
promised, and it was soon annexed by the kingdom of 
Ma'fn (attested from ca. 700 B.C. to 120-100 B.C., 
with its capital Karnaw, Krnw, in the lower Djawf), 
and then some decades before the disappearance of 

Ma'in, by Saba 3 (sec al-Sawda 3 52 = CIH 440, the 
oldest Sabaean language text found at the site). This 
was the time when the Arabs, notably the tribe of 
Amir (W or W") settled in large numbers in the 
Djawf (Robin, 1992, 59-60, 158); a Madhabi text 
from Nashshan, whose ductus dates from around the 
2nd century B.C. (al-Sawda 3 80 = RES 2917 = M 
139), mentions precisely "the war of Amir" (dr ^mr) 
(1. 2). 

At the time of his South Arabian expedition, the 
Roman Aelius Gallus, who reached Marib [q.v.] in 25 
B.C., seems to have taken Nashshan, for amongst the 
towns of the Djawf which the legions took by storm, 



Hanescum) (von Wissmann 1976, 317, 401). Around 
A.D. 80-90, Nashshan was a Sabaean stronghold with 
a royal garrison when the kingdom of Hadramawt in- 
vaded the Djawf during the reign of Karib 3 il Bayan, 
king of Saba 3 and Dhu Raydan, son of Dhamar'alT 
Dharih (Ja 643/22). A 3rd-century inscription shows 
that one of the great Sabaean families, the Banu 
c Uthkulan c Asiyat, had clients (Mm) at Nashshan (Fa 
76/7). Only one inscription from the site of al-Sawda 3 
goes back to this period, al-Sawda 3 51 = 1H 604, a 

ty of the crown's landed property, palm groves, vines 
and fields of various kinds. Nashshan passed under 
Himyaritic control when Himyar [q.v.] annexed 
Saba 3 at the end of the 3rd century, and the last men- 
tions of the place date from the 4th century. The site 
was abandoned between the end of the Himyaritic 
period (6th century) and the 10th century, for the 
Yemeni scholar al-Hamdam [q.v. ] mentions it as a 
ruin (al-khariba al-Sawda^ bi 'l-shakiriyya, ed. Miiller, 
167/12, see also tr. Faris, 1938, 64, 66, Faris, 1940, 
104, 108) belonging to the NashkiyyQn (ed. Khatib, 

Our knowledge of the religion of Nashshan is con- 
fined to two lists of deities and the names of some 
temples. The official pantheon of independent Nash- 
shan, given in two texts (al-Sawda 3 3 and 5), includes 
in the protocolary order c Athtar Sharikan, Wadd, 
Aranyada c , c Athtar dhu Garab and c Athtar Nashk. 
The inscriptions further mention other gods, such as 
c Athtar Matab Khamir, c Athtar dhu Kabd and Rahs 3 
(al-Sawda 3 3 and 5). Four temples are known from the 
texts: Risaf, consecrated to c Athtar (this temple, 
whose name is attested solely in the compound 
c Athtar dhu Risaf, may possibly be identifiable with 
the one excavated by the French Mission outside the 
town); Saywad, the temple of 'Athtar Matab Khamir. 
to be located in the eastern part of the town, according 
to al-Sawda 3 3 which mentions it; Garb um , the temple 
of 'Athtar, attested in the expression "the temple of 
c Athtar dhu Garb unl ", whose building is com- 
memorated by al-Sawda 3 5 = CIH 428 but whose ex- 
act localisation is unknown; and Nasab, the temple of 
Wadd, mentioned in al-Sawda 3 4/2, whose localisa- 
tion is equally unknown. To these temples should be 
added that of Almakah, built "in the centre of the 
town of Nashshan", after the Sabaean victory (RES 
3945/16). Finally, several dedications to 'Athtar dhu 
Kabd from the Minaean period (al-Sawda 3 24-6) lead 
one to think that there was also a temple dedicated to 
this deity. 

The first description of the site of al-Sawda 3 is owed 
to the expedition of Joseph Halevy (a Jew from 
Adrianople, of Hungarian nationality, who became a 
naturalised French citizen after the expedition) and 
his Yemeni Jewish guide Hayyim Habshush, in the 
spring of 1870 (Halevy, 1872, 82-5, 200-14, who 
writes the name as "Es-Soud"; Habshush, ed. and tr. 


Goitein, 45-6). Finally, the site was visited by the 
Egyptians Muhammad Tawflk in 1944-5 (Tawflk, 
1951, pi. 1, marking the itinerary followed) and 
Ahmad Fakhri in 1947 (Fakhri, 1951, i, 147), then by 
the Russian P. A. Griyaznevi^ and the Italian Paolo 
Costa in December 1970 (Griyaznevif, 1978, 220-1), 
and finally by the French Mission from September 
1980 onwards (Robin, 1980, 192-3). 

Bibliography. 1. Inscriptions. CIH = Corpus 
inscriptionum semiticarum. IV. Inscriptions Himyariticas 
et Sabaeas continent, Paris 1899-1930; Fa 76, see 
Fakhry, 1951, ii; la, 643, 647, 664-5, see A. 
Jamme, Sabaean inscriptions from Mahram Bilqis, 
Baltimore 1962; M, see G. Garbini, Iscrizioni 
sudarabiche. I. Iscrizioni minee, Naples 1974; RES = 
Repertoire d'e'pigraphie semitique, Paris 1900-67; al- 
Sawda', see Avanzini; SW/BA/I/1-6, see Breton, 

2. Texts. Hamdani, Si/a, ed. Miiller, Leiden 
1884-91; N.A. Faris, The antiquities of South Arabia 
(= tr. of Book VIII of the Mil), Princeton 1938; 
Faris (ed.), Ar. text of Book VIII, Princeton 1940; 
Book X, ed. Muhibb al-DIn al-Khatib, Cairo 
1368/1948-9; Hayyim Habshush, An account of 

Joseph Halevy 's journey to Najran in the year 1870. . . , ed. 
with summary tr. S.D. Goitein, Jerusalem 1941. 

3. Studies. J. Halevy, Rapport sur une mission ar- 
cheologique dans le Yemen, in JA, 6th ser., xix (Jan.- 
June 1872), 5-98, 129-266, 489-547; Ahmed 
Fakhry, An archaeological journey to Yemen (March-May, 
1947), 3 parts, Cairo 1951-2; Muhammad Tawflk, 
Les monuments de MaHn (Yemen), Cairo 1951 (in 
Arabic); Gabriella Moscati-Steindler, Hayyim Hab- 
sis, immagine dello Yemen, Naples 1976; H. von 
Wissmann, Die Geschichte des Sabarereichs und der 
Feldzug des Aelius Callus, in Hildegard Temorini and 
W. Haase (eds.), Aufslieg und Niedergang der 
Rdmischen Welt. II. Principal, ix/1, Berlin 1976, 308- 
544; A.F.L. Beeston, The transliteration , in Corpus des 
inscrs. et antiquites sud-arabes, i/1, Louvain 1977, pp. 
xxxiii-xxxvi; P. A. Griyaznevic, V poyskakh zateryan- 
nikh gorodov Yemenskie reportazi, 1978, Ger. tr. idem, 
Im Reich da Konigin von Saba, Leipzig 1985; J.-F. 
Breton, Le sanctuaire de '■Athtar dhu-Risdf d'as-SawdP 
(Republique du Yemen), in Acad, des Inscrs. et Belles- 
Lettres, comptes rendus (1992), 429-53; Ch. Robin, Les 
etudes sudarabiques en languefrancaise: 1980, in Raydan, 
iii (1980), 189-98; idem, Inabba\ Haram, al-Kdfir, 
Kamma et al-Harashif, Inventaire des inscrs. sudarabi- 
ques, i, fasc. B, Les planches, Paris-Rome 1992; 
idem, Sheba, in Supplement au Dictionnaire de la Bible, 
to appear; Allessandra Avanzini, as-Sawda. Inven- 
tario delle iscrizioni sudarabische, 3, Paris 1995. 

(Ch. Robin) 
SAWDJ-BULAK, a Persian corruption of soghuk 
bulak "cold spring", Kurdish Sa-blagh, the name of 
a district in southwestern Adharbaydjan, to the 
south of Lake Urmiya, and also the former name of 
its chef- lieu, the modern Mahabad [q.v.]. The 
district comprises essentially Mukri Kurdistan, in- 
habited by the sedentary MukrT and Debokhri tribes 
of Kurds, speaking the KurmandjI form of the Kur- 
dish language (classically described by O. Mann in his 
Die Mundarl der Mukri-Kurden. Kurdisch-persische 
Forschungen, 4th ser. vol. iii/1-2, Berlin 1906-9. Cf. 
Minorsky's bibliography on Kurdish in EI 1 art. 
kurds. E. Language), and confessionally being Sunm 
Muslims of the Shafi c r madhhab. There were formerly 
also Neo-Aramaic-speaking Jews in the town of 
Mahabad. For a detailed geographical and 
topographical description of the Sawdj-bulak district, 
see Minorsky's EI 1 art. s. v. , on which the present arti- 
cle is based. 

The history of the Kurdish tribes of the district is 
only really known from Safawid times onwards. 
During the 19th/16th century, the Mukri chiefs pur- 
sued an opportunistic policy between the rival powers 
of the Ottomans and Safawids. Sarim b. Sayf al-DIn 
challenged and defeated in battle Shah Isma c il I in 
912/1506-7 and then later sought investiture of his 
lands and military help from Sultan Selfm I. His suc- 
cessors variously sought the aid of Shah Tahmasp I 
and Sultan Suleyman the Magnificent. Towards the 
end of the century, Sarim's great-great-nephew 
Amlra Beg (II) adhered to Sultan Murad III, who 
added to his hereditary lands in Mukri Kurdistan the 
wilayet of Shahrizur and the sandjak of Mawsil. But 
after the Ottoman conquest of Adharbaydjan during 
the Perso-Turkish War of 986-98/1578-90, the Ot- 

(as he now was) was made subordinate to the Ot- 
toman governor in Tabriz, reduced to authority only 
over his hereditary lands and compelled to pay an an- 
nual tribute of 15 kharwars of gold. In the early 
llth/17th century, the historian Iskandar Beg Mun- 
shi [q.v.] was an eye-witness of Shah c Abbas I's ex- 
pedition against the Mukri and Bradost Kurds (see his 
Ta^rikh-i c Alam-drd, tr. R.M. Savory, Boulder, Colo. 
1978, ii, 1009-19, years 1018-19/1609-10), and the 
episode of the siege of Dumdum Kal c a (south of Ur- 
miya on the Kasimlu river) became a favourite theme 
of later Mukri heroic ballads. 

In the early 19th century, the Turkish governor for 
the Kadjars in Maragha, Ahmad Khan Mukaddam, 
took strong measures against the Mukris by massacr- 
ing several of their chiefs. In the last decades of the 
century there was fierce sectarian fighting between the 
Sunn! Kurds of Sawdj-bulak and the Shifts of the 
Maragha district. After 1905 the Sawdj-bulak district 
was gradually taken over by Ottoman forces under 
Mehmed Fadll Pasha, until in 1914 the old frontier 
between Turkey and Persia was restored through its 
delimination by Russian and British representatives. 
The region was the scene of fierce Russo-Turkish 
fighting in the First World War, beginning with the 
assassination of the Russian Consul at Sawdj-bulak, 
Colonel A. Iyas, at nearby Miyandoab. 

After the War, under Rida Shah Pahlavi [q.v.] the 
name of Sawdj-bulak town was changed to Mahabad. 
For the subsequent history of the modern town, and 
especially its role as capital of the short-lived Kurdish 
Republic of 1946, see mahabad. 

Bibliography: Sir R. Ker Porter, Travels in 
Georgia, Persia etc., London 1822, iii, 453-98; Col. 
Monteith, Journal of a tour through Azerbaidjan, in 
JRGS, iii, 1833, 5-6; J.B. Fraser, Travels in Kur- 
distan, London 1834; J.C1. Rich, Narrative of a 
residence in Kurdistan, London 1836, i, 223-260; H. 
Rawlinson, Notes on a Journey from Tabriz in 1838, in 
JRGS, x (1840), (extremely important article); Rit- 
ter, Erdkunde, 1838, viii, 393; 1840, ix, 586, 597, 
603-4, 631, 807, 822, 940, 944, 1014-36; M. 
Wagner, Reise nach Persien und dem Land der Kurden, 
Leipzig 1852, ii, 100-2; Scheref-nameh, ed. W. 
Veliaminof-Zernof, St. Petersburg 1860, i, 279-96; 
tr. F.B. Charmoy, St. Petersburg 1873, ii, 135-53; 
Thielmann, Sireifzuge im Kaukasus , Berlin 1875, 321; 
G. Hoffmann, Auszuge aus syrischen Akten pers. Mdr- 
tyrer, in Abh. f d Kunde des Morgenlandes, 1880, 208- 
16; H. Schindler, Reisen im nordwestlichen Persien, in 
the Zeitschr. d. Gesellsch. f Erdkunde zu Berlin, xviii 
(1883), 341-43; Mrs. Bishop (Isabella Bird), 
Journeys in Persia and Kurdistan, in the Geogr. Journal, 
London 1891, ii, 207-10;J. de Morgan, Miss, scient. 
en Perse, Etudes geogr., ii, Paris 1895, 1-44 
(photographs, maps); W.B. Harris, From Batum to 


Baghdad, London 1896, 162-233; idem, A journey in 
Persian Kurdistan, in Geogr. Journ. , iv (1895), 453-60; 
S.G. Wilson, Persian life and customs, 2nd ed., Lon- 
don 1896, 99-122; A. Billerbeck, Das Sandschack 
Suleimania, Leipzig 1898; Ghilan, Les Kurdes persans 
etV invasion ottomane, in RMM (1908), 1-22, 193-210; 
C.F. Lehmann-Haupt, Armenien einst und jetzt, 
Berlin 1910, i, 219. See also the maps and some 
photos in E. Herzfeld, Paikuli. Monuments and in- 
scriptions of the early history of the Sasanian empire, Berlin 
1924. (V. Mi_norsky-[C.E. Bosworth]) 

SAWDII, SAWDlI, the name of three 
Ottoman princes. 

The name would appear to originate in the Old 
Turkish (especially, Eastern Turkish) word saw 
"word, piece of discourse, utterance", found as early 
as the Orkhon inscriptions, then in the Turfan 
Uyghur texts, in the late 5th/llth century Kutadghu 
bilig [q.v.] and up to the 8th/14th century, after which 
it is not attested as a separate word (Clauson, An 
etymological diet, of pre-thirteenth century Turkish, 782-3). 
Cf. also the name of the slave commander of the 
Saldjuk sultan Alp Arslan, Sawtigin. Sawdji would ac- 
cordingly be "purveyor/conveyor of a message > 
prophet". In the later 8th/14th century it appears in 
onomastic in Sawdji Agha, head of a contingent of 
troops in Murad I's army against the Karamanids, 
and it is about this time that it is attested in the Ot- 
toman royal house itself (see below). 

(1) Sawdji Beg, in the old Ottoman chronicles also 
called Sari yati or Sar! balI, etc. was one of the 
younger brothers of c Othman [q.v.], the founder of 
the Ottoman dynasty, and a son of Ertoghrul. He 
supported his brother on his campaigns and fell 
(684/1285-6 is the date given) in battle against the 
governor of Angelokome (Ayne G61) at Egridje south 
of Koladja, behind Olympus at the foot of a pine tree. 
The tree was still called Kandili cam "pine tree of the 
lamps" in later times presumably from the lights lit 
there, the glimmer of which was afterwards given a 
mystic significance. (According to Neshri, Idrls 
BidllsTand Sa c d al-Din, Tddfal-tawankh, i, 18,8 ff., a 
heavenly light, nuzul nur, illuminated the tree by 
night.) Sawdji Beg was buried beside his father in his 
tomb (turbe) at Sogiid [q.v.) in northwestern Anatolia, 
destroyed by the Greeks in 1922. 

Bibliography:], von Hammer, GOR, i, 54, and 
following him J. Zinkeisen, GOR, i, 70 (from Sa c d 
al-Din, who follows Idrls Bidllsl, Hesht bihisht, and 
Neshri, Djihdn-numd). 

c Othmc 

(3) The eldest son of Murad I [q.v.] who, when 
governor of Rumelia, made terms with a son of John 
V Palaeologus of Byzantium named Andronicus, and 
rebelled against his father. The Ottoman chroniclers 
give very scanty information about this conspiracy, 
while the Byzantine historians Chalcocondyles, 
Phrantzes and Ducas give very full accounts, differing 
only in details; cf. Chalcoc, ed. Imm. Bekker, i, 
40 ff. (Eaoufo); Phrantzes, ed. Bekker, i, 50, where 
the rebel is wrongly called Moot] T^ikt-KTU, i.e. Musa 
Celebi through confusion with Bayezid I's son; 
Ducas, ed. Bekker, 22 (Eotpoutiiio,;), where Sawdji is 
mentioned but the rebel is called Kouvtou^t);, i.e. 
Gundiiz. Murad I acted jointly with John V and took 
the field against the two princes. After an unsuccessful 
battle at a place which the Byzantine writers call 
'AnixpiSiov (Chalc, 43,4), Sawdji fled to Didymo- 
tichon, where he was surrounded and forced to sur- 
render to his father. He was blinded and then behead- 

ed. The execution took place in 787/1385-6 and the 
body was brought to Bursa and buried there. Murad 
I had apparently made up his mind to get rid of 
Sawdji, as he had appointed his son Bayezid to watch 
his movements, cf. Murad I's letter to Bayezid in 
Feridun, Munshe^at-i selatin, i 2 , 107 (of the beginning 
of Rabi< I 787/1385-6), with Bayezid's answer, op. 
at., esp. 108 above, according to which the Kadi of 
Bursa must have passed a death sentence on Sawdji. 
The execution of Sawdji was the first of a long series 
of similar cases, in which princes dangerous to the 
Ottoman heir-apparent were put out of the way. 
Bibliography: J. von Hammer, GOR, i, 190, 
599; Zinkeisen, in GOR, i, 237 ff.; Hadjdji Khalifa, 
Takwim al-tawdrikh, under the year 787; Sa c d al- 
Dln, Tidxal-tawirikh, i, 100 (following Idrls Bidllsl); 
A.D. Alderson, Structure of the Ottoman dynasty, Ox- 
ford 1956, index at 157; tA art. Sava (M. Tayyib 
Gokbilgin). (F. Babinger-[C.E. Bosworth]) 
SAWIK (a.) is a food preparation of some antiqui- 
ty, and one widely known throughout the mediaeval 
Middle East. Al-Tha c alibi attributes its first ap- 
pearance to Alexander the Great, and it is cited in the 
physicians' works of both eastern and western Islamic 
lands. It was recommended for travellers and was us- 
ed to feed armies in the field. For all its fame, it rarely 
appears described in the extant culinary manuals, 
although some recipes are found in the earliest 
(4th/10th century) work by al-Warrak (K. al-Tabikh, 
ed. K. Ohrnberg and S. Mroueh, Helsinki 1987, 37- 
8). Preparation was chiefly from wheat and barley, 
the former preferred among the urban classes which 
could afford it. The wheat grain was first washed and 
then soaked in water overnight. Discarding the water, 
the wheat was next fried thoroughly until browned. 
When cooled, it was ground, sieved and then stored 
for use when it could be eaten by adding sugar. An 
alternative, more complicated method was to husk 
and dry the grain before frying. This basic prepara- 
tion could then be used in other types; for example, 
in sawik rumman three portions of wheat sawik to one 
of pomegranate (rumman) seeds were mixed together, 
cooked, sieved and sugar added. Sawik was also added 
to the dough in making the pastry, ka}k. According to 
al-Razi (K.fidaf muddrr al-aghdhiya, Cairo 1888, 7), 
pomegranate and apple sawik were intended only for 
medicinal purposes, while wheat and barley types 
were for nourishment. Plain sawik was considered a 
nourishing substitute for fresh fruit when it was 
unavailable. Medicinal preparations made from 
barley are described by Isaac Israelii (K. al- aghdh iya. 
facs. ed., Frankfurt 1986, ii, 72-8), and clearly physi- 
cians envisaged different ways in which it could be 
used to achieve different effects within a regimen of 

Bibliography: In addition to the above 
references, see Kanz al-fawaHd fi tanwf- al-mawaHd, 
eds. M. Marin and D. Waines, Beirut-Stuttgart 
1993. (D. Waines) 

SAWM (a.), a term of Islamic law, denoting the 
bargaining involving both vendor and purchaser 
that occurs before a sale. Sawm is a classical term 
which, although pre-contractual, influences the for- 
mation of the contract and has a legal effect upon it. 
Sale is prohibited if a higher bid is offered by a third 
party during the negotiation leading to the sale agree- 
ment. Al-BukharT and Muslim record the prohibition 
of making an offer while another's offer is being con- 
sidered. What is curious is that they also record the 
prohibition of sale (bay'). This has lead to some confu- 
sion among scholars. The difference between sawm 
and bay' is that the former is no more than an offer to 


enter into the latter after the manifest approval of the 
vendor. The prospective purchaser can take the com- 
modity home for examination, even if he proposed a 
lower price than the vendor requested. If, during this 
period, the commodity is damaged, then the full price 
requested by the vendor becomes liable. It is in- 
teresting to compare bay'- and sawm with khitba and 
nikdh [q.v.] since they represent parallel sequences. 
Hadith often refers to the two types of contract 
together. This indicates the personal aspects of bay c in 
Muslim society as well the formal dimension in 

Bibliography: Abu 'l-Walld al-Kurtubl, al-Baydn 
wa 'l-tahsil, ed. M.S. A c rab, Beirut, vii, 486-7; L'A, 
Beirut, bar Sadir n.d., xii, 310; Ibn Hadjar al- 
'Askalani, Fath al-Bdri, iv, 325-53; V-A, Cairo 
1306, viii, 350; Shafi c i, K. al-Umm, Beirut 
1403/1983, iii, 92; Ibn Kudama, al-Mughm, ed. 
Hulw and TurkI, Cairo 1408/1988, vi, 305-8. 

(M.Y. Izzi Dien) 
§AWM (a.), with Siyam, masdar from the root 
s-w-m; the two terms are used indiscriminately. The 
original meaning of the word in Arabic is "to be at 
rest" (Th. Noldeke, Neue Beitrdge zur sem. Sprachw., 
Strassburg 1910, 36, n. 3; see previously, S. Frankel, 
De vocab. ... in Corano peregrinis, Leiden 1880, 20: 
"quiescere"). The meaning "fasting" may have been 
taken from Judaeo-Aramaic and Syriac usage, when 
Muhammad became better acquainted with the in- 
stitution of fasting in Medina; this is the sense of the 
word in the Medinan suras. 

Origin of the rite. That fasting was an unknown 
practice in Mecca before Muhammad's time cannot 
be a priori assumed. Why should not the hunafa? [see 
hanif] have also used this religious custom? In the 
Meccan suras, as above mentioned, there is a 
reference to sawm in XIX, 27: a voice commands 
Mary to say "I have made a vow of sawm to the Mer- 
ciful, wherefore I speak to no one this day". Observ- 
ing silence as a Christian fasting practice (cf. Afrahat, 
ed. Parisot, in Patrol. Syriaca, i, 97) may have been 
known to Muhammad. In the year 2/623-4, according 
to unanimous reliable Muslim tradition (cf. A.J. 
Wensinck, Mohammed en de Joden te Medina, Leiden 
1908, 136-7, contra e.g. A. Sprenger, Das Leben und die 
Lehre des Mohammad, iii, 53-9), the revelation of sura 
II, 1 79-81 , abolished the 'Ashura* fast as an obligation 
by the institution of the fast of Ramadan. On the 
question why Muhammad chose this particular 
month, various opinions have been expressed. The 
most plausible is that of A.J. Wensinck, who has 
called attention to the particularly sacred character of 
the month of Ramadan even in pre-Islamic times (on 
account of the laylat al-kadr in his Arabic New-Year and 
the Feast of Tabernacles, in Verh. Ak. W. Amst., N.S. 
[1925], xxv/2, 1-13; see also M.Th. Houtsma, Over de 
Israelitische Vastendagen, in Versl. en Med. Ak. Wetensch., 
Afd. Letterk., Series 4, ii [Amsterdam 1898], 3 ff). 
The first regulations concerning the manner of 
the Muslim fasting are given in sura II, 179-81 , which 
probably belong together (Noldeke-Schwally, 178, 
contra Th.W. Juynboll, Handbuch des isldmischen 
Gesetzes, Leiden- Leipzig 1910, 114, who considers v. 
181a later revelation; al-Baydawi also assumes that it 
was revealed in separate parts). The fast is also men- 
tioned elsewhere in Kur'an II, 192; IV, 94; V, 91, 96; 
LVIII, 5. Sd>im is further used in XXXIII, 35, to de- 
scribe the devout Muslim, along with other epithets, 
while in II, 42 and 148, sabr [q.v.] is explained as 

were taken from Tradition. What follows here is a 
resume of the law on fasting according to the Shafi'I 
school, as contained in the treatise by Abu Shudja c al- 
Isfahanl (5th/llth century), al-Mukhtasar fi 'l-fikh. 

How the fast should be observed and who is 
boundtofast. Fasting in the legal sense is abstinence 
(imsdk) from things which break the fast (muftirdt), 
with a special niyya (intention). The s<Pim must be a 
Muslim, in full possession of his senses ( c dkil), and, if 
a woman, free from menstruation and the bleeding of 
childbed. The fast is valid (sahih) under these condi- 
tions; there is an obligation to fast on every one who 
is full-grown (bdligh) if he is able physically to sustain 
it (kddir). 

One ought to formulate the niyya before dawn on 
each day of fasting (tabyit); by talfik [q.v.], however, 
the Shafi'I can follow the Malik! madhhab, which 
allows one to formulate the niyya for the whole of the 
month of Ramadan. 

The muftirdt are defined by the entry into the body 
of material substances, in so far as they are material 
in substance and can be prevented from entering. In 
order for there to be iftar, there must be deliberate ac- 
tion (ta'ammud), knowledge (Him) and free choice (ikh- 
liydr) (al-Bukharl, Sawm, bdb 26, Aymdn, bdb 15; 
Muslim, Siyam, tr. 171). 

It is recommended for the s&Hm to take the fatur 
(meal marking the end of the fast) as soon as he is sure 
that the sun has set, and the sahur (meal taken after 
midnight) as late as possible; to abstain from indecent 
talk and calumnies; to avoid actions which might 
stimulate the passions within one or in others; to recite 
the Kur 3 an for oneself or for others; and to observe 
retreat (Hikdf [q. v. ]) during the month of Ramadan. 
Al-Ghazall adds to these the duty of charity towards 

I. Obligatory (wddj.ib) fasting. 

Fasting in the month of Ramadan is the fourth 
pillar of Islam; whoever denies the obligation to fast 
is a kdfir, except for the recent convert or one who has 
not been in contact with the ( ulamd\ Whoever omits 
to fast without good cause, without, however, denying 
the compulsion to fast, may be imprisoned. The 
general obligation to fast ('aid sabil al-'umum) begins 
on 1 Ramadan, after 30 Sha'ban, or after 29 if the 
hakim (kadi) has then accepted the evidence of one c adl 
(sc. a person worthy of credence) that he has seen the 
new moon. The beginning of Ramadan may be an- 
nounced to the people in a way settled by the local 
custom (a cannon, the hanging of lamps on the 
mandra, in Java by beating the bedug). Days omitted in 
Ramadan have to be made good (kadd 7 ) as soon as 
possible, but not on one of the forbidden days or on 
one which is itself a compulsory fast day. 

A distinction is made between the major and minor 
kaffdra (compensating). The first is imposed on 
anyone who (a) breaks the fast in Ramadan by sexual 
intercourse; (b) is guilty of illegal killing [see katl]; 
(c) has pronounced the zihdr formula but not the taldk 
immediately after it; (d) has broken a valid oath 
(yamin) [see kasam]. 

The minor kaffara or fidya has to be paid when one 
takes advantage of one of the dispensations which are 
detailed below; the question of fasting does not arise. 

In the case of great drought, the Imam may, accord- 
ing to the shari'a, prescribe extraordinary ceremonies 
which include fasting; the three days before the saldt 
al-istiskd'' [see istiska 3 , and al-Badjuri, Kitdb Ahkdm al- 
saldt, fast fi ahkdm saldt al-istiskd 1 '] are spent in fasting. 

The law permits relaxations in the following cir- 

n age (men 40; r 

(a) Such as have reached a 


exactly defined for women) and sick people for whom 
there is no hope of recovery, if they are unable to fast, 
may omit the fast without being bound to the kadP 
should their strength or health be restored. In com- 
pensation they should give alms at the rate of one 
mudd for each day omitted. 

(b) If pregnant or nursing women fear it would be 
dangerous for them if they should fast, iftar is wddjib 
or obligatory for them but W« ! is obligatory. 

(c) Sick persons who are likely to recover and those 
who are overcome by hunger and thirst may break the 
fast on condition that the kadd^ is performed. If a man 
is in danger of death or danger of losing a limb, iftar 
is wddjib. 

(d) Travellers who set out before sunrise may, if 
necessary, break the fast, but not if they begin their 
journey during the day. In case of mortal danger, iftar 
is wddjib. Two days' journey is the minimum. Kadd* 
is obligatory on them, in this case. 

(e) Those who have to perform heavy manual 
labour should formulate the niyya in the night, but 
may break the fast if need be. 

When the justification for relaxing the rules disap- 
pears, it is surma to pass the rest of the day fasting. 
II. Voluntary or supererogatory fasting 


This falls into the category of what is recommended 
(mandub ilayhi). For a woman, it may only be done 
with the consent of her husband. It may be broken 
without any penalty. The niyya, which can be for- 
mulated any time up till midday, need not be explicit- 
ly made, although certain of the fuk.aha' consider it 
desirable in certain cases. 

It is recommended for anyone who has to fast (as a 
substitution) for three days during the hadjdj and 
seven days afterwards, to choose as the three days the 
7, 8 and 9 Dhu '1-Hidjdja. 

For the supererogatory fasting of six days in 
Shawwal. it is preferable to choose the six consecutive 
days immediately after the festival, i.e. 2-7 Shawwal 
(see Juynboll, Handbuch, 132). 

The following days are further recommended for 
voluntary fasting: the day before '■dshurd > day and the 
one after; the yawm al-mi c rddj (27 Radjab); and the 
Monday and the Thursday (surma mu c akkada, accord- 
ing to al-Badjurt), days when actions are offered to 
God, and others which are describable as "by way of 

III. Fasting is forbidden (hardm) on the days of the 
two great festivals, on the tashrik days and for a 
woman during menstruation; also in definite cases 
when danger threatens, as already mentioned above. 

IV. It is blameworthy (makrdh) to fast on Friday 
because it distracts the attention from the Friday wor- 
ship, and on Sunday or Saturday, at least if one has 
no particular reason for fasting, because the Christ- 
ians and Jews observe these as holy days. 

The three other madhdhib diverge from the Shafi'T 
one on minor questions only which it would be tedious 
to enumerate (see c Abd al-Wahhab al-Sha'ranl, K. al- 
Mizdn, Cairo 1279/1862-3, ii, 20-30). 

The Shi 11 ! prescriptions diverge from the Sunnf ones 
on some points of detail, of which the following are 
some examples (see A. Querry, Recueil de lois concernant 
lesMusulmansSchyites, Paris 1871-2, i, 182-209, ii, 75- 
7, 197-9, 203-5, following Nadjm al-DTn al- 
Muhakkik, ShardV al-Isldm fi masdHl al-haldl wa 'l- 
haram): the niyya is not considered as a "pillar"; smo- 
king is not considered as one of the mufiirdt; it is for- 
bidden to blaspheme against the Sh/I Imams; the 
man and woman guilty of sexual relations during the 
Ramadan fast are to be flogged for the first offence, 

with the death penalty for persistent offences. 
Regarding the relaxations, fasting is not required 
for a sick person unless authorised by the physician; 
dispensation is only given to pregnant women if they 
are in final phase of pregnancy; fasting is not general- 
ly required of travellers, but compensation must be 

Voluntary and supererogatory fasting may begin 
before midday. ShlSsm adds to the days recommend- 
ed by the Sunnis the great dates of Shi^i tradition 
(ghadir Khumm, mubdhala, anniversary of al-Husayn's 
death on 10 Muharram; see Querry, op. cit. , 37, 

Al-GhazaTT gives, at the beginning of his Kitdb Asrdr 
al-sawm in the Ihyd?, some considerations on the value 
of fasting. He points out, referring to some well- 
known traditions, the high esteem in which fasting 
stands with God. He gives as a reason for this that 


of defeating the ei 
God, because human passions, which are Satan's 
means of attaining his ends, are stimulated by eating 
and drinking. Fasting is therefore "the gateway to the 
service of God." After having enumerated, in the 
manner of afakih, the regulations regarding fasting, 
he declares that these do not constitute what is essen- 
tial. He distinguishes three steps in the fast. The first 
step is that of the fikh, the third that of the Prophets, 
the siddikun and those who have been brought into the 
proximity [of God] (al-mukarrabun), whose fast 
separates them from all worldly desires. But the 
second step suffices for the pious; it consists in keeping 
one's organs of sense and members free from sin and 
from all things that detract from God. Subjection of 
the passions is the real object of fasting, mot mere 

The ethical conception of the fast which al-Ghazali 
gives in this second fast supplements, he says, the bar- 
ren law of thefukahd\ In the Hadith we find already 
various traditions with ethical tendencies and al- 
GhazalT does not fail to quote them in support of his 
view. A mass of traditions on fasting, grouped accord- 
ing to the various aspects of the topic, will be found 
in Wensinck's Handbook of early Muhammadan tradition, 
71-6, s.v. Fast, Fasting. 

It is a very widespread opinion today that fasting, 
especially during Ramadan, forms the most ap- 
propriate expiration for offences committed during 
the year. This is why the law on fasting is still in 
general strongly observed, in varying degrees of 

Bibliography (in addition to works cited in the 
article): A sketch of the legal prescriptions accord- 
ing to the Shafi'T law school was given by Th.W. 
Juynboll in his Handbuch des islamischen Gesetzes, 
Leiden-Leipzig 1910; see the bibl. to the 1925 ed. 
The main sources are the chs. on fasting in the col- 
lections of hadith, fikh and ikhtildf, to which should be 
added Ghazali, Ihyd\ ch. on sawm. For the tradi- 
tions, see Wensinck, op. cit., and idem, Concordances 
et indices de la Tradition musulmane, Leiden 1936 ff. 
See also K. Wagtendonk, Fasting in the Koran, 
Leiden 1968; K. Lech, Geschichte des islamischen 
Kultus. i. Das ramadan- Fasten, Wiesbaden 1979; and, 
above all, the bibls. to hadith. hadjdj, mut c a, 
puasa in Suppl., ramadan and C UMRA. 

(C.C. BERG-[Ed.]) 

SAWTIYYA (a.), the neologism used for the 
modern phonetic description of Arabic. 

1. Definition and background. The sound system of 
Arabic is described on the basis of its phonetics, '■Urn 
al-aswdt, which focuses on the investigation of how 

sounds are produced by speakers and perceived by 
listerners and on its phonology, Him al-sawtiyydt. The 
word sawtiyya is a relative adjective derived from the 
verbal noun, sawt (pi. aswat), meaning sound or 
speech sound. The phonological analysis renders a 
description of the processes that set sounds into 
organised and systematic patterns. These are gov- 
erned by the phonological rules operating in the 
system that make the distinction between the sounds. 
The phonemic inventory of Arabic is established, not 
only on the basis of detailed phonetic information, but 
also on the patterning of the sounds. 

As a result of the impact of the Kur'an on Arabic, 
the literary language has maintained continuity in its 
sound system and grammatical structure. The Arabs 
and Muslims have rendered a meticulous investiga- 
tion of this system. Their original contributions, in 
theory and practice, result in precise phonetic descrip- 
tions and terminology. Pioneers in this field are such 
scholars as al-Khalil and his disciple Sibawayh [j.»»]. 
However, most linguistic changes have occurred with- 
in the dialects. The description of sawtiyya here focuses 
on the modern al-fusha (literary Arabic) sound system. 

2. Description of the sound system. There are 29 con- 
sonants and six vowel phonemes or distinctive sound 
units in Arabic. Consonants are produced with either 
partial or complete obstruction of the airstream in the 
vocal tract and are described by their manner and 
points of articulation. The vocal cords and their func- 
tion also play a definite role in the classification of the 
phonemes into voiced or voiceless. Sounds produced 
with the vocal cords vibrating are called voiced while 
those produced without vibration are voiceless. 

The major subgroups of consonants are: (1) Stops: 
These sounds involve a closure of the airstream at 
some point in the oral cavity. The stops are: /b/ voic- 
ed bilabial; hi voiceless dental; It/ emphatic voiceless 
postdental; /d/ voiced dental; Id/ emphatic voiced 
postdental; /k/ voiceless velar; /q/ voiceless unas- 
pirated uvular. (2) Fricatives: These sounds are pro- 
duced with a partial closure of the oral cavity, where 
the airstream passes through a narrow passage, 
creating friction or a hissing sound. The fricatives are: 
HI voiceless labiodental; /6/ voiceless interdental; Idl 
voiced interdental; lijl emphatic voiced interdental; 
Is I voiceless dental hi emphatic voiceless postdental; 
hi voiced dental; l\l voiceless palatal; /%/ voiceless 
velar; ly I voiced postvelar. (3) Affricate: There is only 
one affricate I3I '. This sound, phonetically, is compos- 
ed of two phones — a stop [d] immediately followed by 
a fricative "[3] like" sound. (4) Nasals: In the produc- 
tion of the nasals, Iml and In/, two cavities, oral and 
nasal, are involved and it is this unique combination 
that distinguishes nasals from other sounds. The Iml 
is voiced bilabial nasal and the In I, voiced dental 
nasal. (5) Trill: The sound hi is a trill voiced alveolar 
or post-dental. (6) Lateral: The sound l\l is a lateral 
voiced dental. The hi and III, when occurring in the 
vicinity of the vowels /a/ or /aa/, may be characteris- 
ed as emphatics. (7) Semivowels: lv/1 is a bilabial 
semivowel and lyl a palatal semivowel. They form 
two diphthongs: /ay/ and /aw/ as in bayt (house) and 
thawb (dress). (8) Pharyngeals: The sound c ayn /7 is 
described as a voiced fricative pharyngeal, however, it 
is also described as a voiceless stop as pronounced in 
'Irak and the Arabian Peninsula. The sound lb. I is a 
voiceless pharyngeal. (9) Glottals: The sound /h/ is a 
voiceless oral fricative and the hamza /'/, a glottal 

3. Emphatics. The feature of emphasis is traditional- 
ly called tafkhim ("thickness"). There are four era- 

d, s, 61. These sounds, in addi- 

tion to their primary points of articulation, are 
characterised by a secondary articulatory feature 
identified as the retraction of the tongue as well as the 
contraction of the pharynx. This phonetic 
phenomenon was recognised by the earliest Arab 
grammarians and has always been a point of attrac- 
tion and fascination for Western linguists and 
Arabists. Emphasis is not limited only to this par- 
ticular group as there are other consonants, on a 
marginal basis, which demonstrate emphasis or 
pharyngealisation. Emphatics are found in modern 
Arabic dialects, to varying degrees, depending on 
geographical areas and social levels of speaking. The 
emphatic consonants condition neighbouring sounds, 
and often dominate the whole syllable. This condi- 
tioning is referred to as "spreading", which can be 
either progressive or regressive and may even cover a 
syllable or more. The acoustic effect of the emphatics 
is to lower the frequency of neighbouring sounds. The 
dotted line of the shape of the tongue on the diagram 
below shows the postion of the tongue for the em- 
phatic hi as in the syllable hi I that contrasts with the 
broken line that represents the non-emphatic hi as in 
Iti I. 

The sound 161 exhibits some changes in its pron 
tion. It is predominantly pronounced as Idl , 
terdental emphatic fricative, in 'Irak and Arabia. 
However, in the sedentary Levant regions, it is pro- 
nounced as /z/. Likewise, the sound Idl is pronounced 
either as 161 in c Irak and Arabia, or as Idl in sedentary 
regions (see further on this, dad). 

nasal cavity 

This diagram is a tracing of a frame from an Xray 
sound film of the author's vocal tract. It is superim- 
posed over a diagram of the vocal tract made by the 
phonetician Peter Ladefoged. 

4. Vowels. Vowels are produced with the a 
passing fairly freely through the oral cavity. There are 
two types of vowels, long and short. The short are: 
kasra HI, short high front unrounded vowel; damma 
lul, short high back rounded vowel; and falha hi , low 
short central unrounded vowel. The long vowels are: 
Hi/ kasra + ya>, long high front unrounded; /uu/ dam- 


ma + waw, long high back rounded and /aa/ fat ha + 
alif, low long central unrounded. Length is phonemic 
or functional and may be indicated by writing the 
vowels twice as in the following minimal pairs: kataba 
("he wrote") and kaataba ("he corresponded with"). 
The difference between long and short vowels is 
primarily one of quantity. 

5. The syllable and its structure. The structure of the 
syllable is based on the main components of its forma- 
tion. Every syllable begins with the onset, a single 
consonant, and may end with a coda, which can be 
either one or two consonants or none. The nucleus or 
centre of the syllable is the most prominent element 
and is always a vowel, short or long. There are six 
possible syllable types. By postulating a "C" 

s the 

syllable types are: CV, CVV, CVC, CVVC, CVCC 
and CVVCC. These are divided into short CV; long 
CVC, CVV; CVVC and heavy CVCC and 
CVVCC. With respect to the frequency of occurrence 
the first four syllable types, CV, CVC, CVV and 
CVVC, are considered the dynamic force behind the 
formation of all the grammatical phonological pat- 
terns of non-pause utterances. The last two types, 
CVCC and especially CVVCC, have very limited 
distribution. They occur only in words, phrases and 
sentences in final pause forms, as does CVVC. 

Consonant clusters occur medially and finally. All 
syllables and words start with a single consonant. In 
the Arabic writing system whenever a word starts with 
an alif, the alif, which is used as a kursT ("chair" for 
the hamza), has no phonetic value. Consonants that 
are produced in the front section of the oral cavity, 
form bilabial to palato-alveolar, form clusters with the 
remaining back consonants. Restrictions exist on the 
formation of consonant clusters, primarily among 
consonants that have the same manner or adjacent 
points of articulation. Gemination involves the 
clustering of identical consonants. Lengthening of the 
vowels does not create vowel clusters, as it is only a 
quantity difference that results in phonemic length for 
the long vowels. 

6. Stress. Stress placement rules are determined by 
the loudness or prominence of a specific syllable 
relative to other syllables within the utterance. 
Loudness is not the sole feature responsible for stress; 
length and pitch also contribute. Syllable structure, 
types and their distribution play a definite role in the 
placement of stress. The word-stress rule patterns that 

a string of CV types the first syllable is stressed as in 
dd-ra-sa, CV-CV-CV, "he studied", (2) when a word 
contains only one long syllable that syllable (but not 
the final one) is stressed as in mu-ddr-ri-su-hu, CV- 
CVC-CV-CV-CV, "his teacher"; (3) when a word 
contains two or more long syllables the long syllable 
nearest to the end (but not the final one) receives the 
stress as in mu-dar-ri-sda-tu-kum, CV-CVC-CV-CVV- 
CV-CVC, "your teachers" (fern.); and (4) final 
heavy syllables are stressed as in kt-tdab, CV-CVVC, 
"book". These stress rules encounter some variation 
depending on the geographical region that an Arabic- 
speaking person comes from. 

Bibliography: For a description of the Arabic 
sounds of the classical period, see the Bibl. to 
makharidj al-huruf. P: Ladefoged, A course in 
phonetics, 3 Fort Worth 1993; Salman al-Ani, Arabic 
phonology: an acoustical and physiological investigation, 
The Hague-Paris 1970; idem (ed.), Readings in 
Arabic linguistics, 2 Bloomington, Ind. 1992; Ibrahim 
AnTs, al-Aswat al-lughawiyya, 3 Cairo 1961; Tamam 
Hassan, al-Lupha al- c Arabiyya ma'-ndhd wa-mabndha, 

Cairo 1973 (esp. chs. 2-3); S. Davis, On the 
phonological emphasis in two modern Arabic dialects, in 
Internal. Jnal. of Islamic and Arabic Studies, viii/2 
(1991), 1-20. (Salman H. al-Ani) 

SA'Y (a.), from the root s-<-y, used 30 times in the 
Kur'an in such senses as "to work, apply oneself to, 
denounce, seek to earn one's living, run after s. th." 
etc., but in the sense concerning here denoting the 
pilgrim's running between al-Safa and al-Marwa. 
These are two hills to the south and north-west of the 
Ka'ba respectively, linked by a mas% course, which 
the pilgrim follows after having made the sevenfold 
circuit of the Ka c ba, at his or her arrival and his or 
her departure. This following of the course, the sa'-y, 
is likewise sevenfold; it starts in al-Safa, and goes to 
al-Marwa, ca. 300 m away. In the first part of the 
course is a smaller course of about 80 m, marked out 
by four green columns, which the men — but not the 
women — have to make whilst running, the remainder 
of the course being made at a normal walking pace. 
From al-Safa four trips are made to al-Marwa, and 
from this last three trips back are made, the seventh 
trip ending at al-Marwa, the whole amounting to a 
little more than two km (see M. Hamidullah, Le 
pelerinage a la Mekke, in Sources orientales, iii, Paris 1960, 

Sa'-y is the term reserved for this travelling along the 
course, the other ones (the tawaf around the Ka c ba 
and those between c Arafa-Muzdalifa-Mina) being 
termed fayd or ifada (course made in an enthusiastic 
manner). It is obligatory at arrival and superogatory 
at departure for the pilgrims who perform both the 
Greater (ha djdf ) and Lesser (himra) Pilgrimages [see 

The first Muslims, and especially the Ansar, 
hesitated to make this sacred course, because of its 
pagan character. On the two hills were the idols Isaf 
and Na'ila [q.v.], two sacred stones, on which the vic- 
tims' blood was poured. They must have had a vague- 
ly human form which, under Syrian influences and 
because of their proximity to each other, were 
assimilated to a divine pair, as with a Ba c l and a Ba c la. 
But an edifying legend developed round them, that 
they were a man and woman who had fornicated in 
the Ka c ba and had therefore been transformed into 
stone (see T. Fahd, Le pantheon de I'Arabie Centrale a la 
veille de I'he'gire, Paris 1968, 103-9). 

The question of this obligation of the sa'-y was put 
to the Prophet, and the reply was revealed that "al- 
Safa and al-Marwa are part of the rituals (sharaV) of 
pilgrimage to the House of God. Whoever ac- 
complishes it (hadjdja) or visits it (ftamara) will never 
commit a sin, if he makes the course between the two 
hills" (Kur'an, II, 158). Some recent constructions 
have been made to facilitate the sa'y, involving two 
superimposed tracks, one for the outward journey and 
one for the return (see the monthly journal al-'-Arabi, 
Kuwait, no. 1971). 

Bibliography: For further details on the rite, see 
M.M.I. c Abduh, al-Hbaddlfi 'l-Isldm, Cairo 1954, 
420-4; Hamidullah, op. at., 1 1 1. For the rites of the 
Pilgrimage in general, see Fahd, Le pelerinage a la 
Mekke, in Etudes d'Histoire des Religions, i, Paris 1974, 
65-94; idem, Pantheon, esp. 241-7, but the most 
complete work is M. Gaudefroy-Demombynes, Le 
pelerinage a la Mekke, Paris 1923. (T. Fahd) 

SAYABIglA, for Sayabiga, the name given in ear- 
ly Islamic historical sources to a group of non-Arab 
emigrants, proximately from Sind in India but, 
most probably, ultimately from South-East Asia and 
established on the Arab shores of the Persian Gulf and 
at Basra in the first two centuries or so of Islam. 


Arabic authors often link them with the Zutt [q.v.] or 
Jhats [see djat ] from northwestern India (see e.g. al- 
Tabari, i, 1961, 3125, 3134, 3181), although two 
distinct ethnic groups are in fact involved here. 

De Goeje was the first to discuss the Sayabidja at 
length, in his Memoire sur les migrations des Tsiganes a 
havers VAsie, Leiden 1903, 18-19, 86-91, in the latter 
place giving a resume of an earlier article of his 
devoted to them. He noted that, already in pre- 
Islamic times, there are mentioned Sayabidja settled 
on the Gulf coastlands and Zutt in lower c Irak (al- 
Baladhurl, Futuh, 373), and that in Abu Bakr's time 
there was a garrison of Sayabidja and Zutt at al-Khatt 
[q.v. ] along the coasts of al-Katlf and Hadjar (al- 
Tabari, i, 1961). Both these peoples must have been 
planted there, or induced to settle there, as guards and 
frontier auxiliaries, by the Sasanid emperors. At the 
opening of Islam, both groups followed the example of 
the Asawira or cavalrymen of the Sasanid army (see 
C.E. Bosworth, EIr, art. Asawera) and became 
Muslims on various favourable conditions; the Zutt 
and Sayabidja attached themselves as mawali [see 
mawla] to the Arab tribe of Hanzala of TamTm, and 
Abu Musa al-Ash c an [q.v.] then settled them at Basra 
(cf. al-Baladhurl, Futuh, 373-5; al-Tabarl, i, 2562 ff.). 
Other Sayabidja elements seem to have found their 
way to Kufa and possibly elsewhere, since the Kufan 
forces which joined C A1I at Dhu Kar in 36/656-7 in- 
cluded a detachment of Zutt and Sayabidja (al- 
Tabari, i, 3180-1). 

Sayabidja troops fought in Khurasan under c Abd 
Allah b. c Amir [q.v.] against the remnants of Persian 
resistance there and, in the time of al-Hadjdjadj b. 
Yusuf [q.v.], some of them were to be found in the 
ranks of Ibn al-Ash c ath's [q.v.] rebel forces (al- 
Baladhurl, 374). The caliph Mu c awiya had further 
transplanted Zutt and Sayabidja in the thughur [q. v. ] of 
northern Syria and at Antioch (loc. cit.). Within the 
misr of Basra, a corps of 40 or 400 Sayabidja under 
one Abu Salima al-Zutti is mentioned as guarding the 
state treasury there in 36/656-7 (al-Baladhuri, Futuh, 
375-6; al-Mas c udI, Murudj, iv, 307 = § 1629, and cf. 
Pellat's Index generaux, vi, 403), and a verse in a poem 
of the Basran poet Ibn al-Mufarrigh (d. 69/689 [q. v. ]) 
mentions "the uncouthly-speaking Sayabidj bar- 
barians who loaded me with fetters in the morning" 
(Ibn Kutayba, Shi c r, 212). It would accordingly ap- 
pear that the Sayabidja were a rough and tough ele- 
ment who were employed on police and custodial 
duties. The Sayabidja of Basra are mentioned at the 
time of Ibn al-Zubayr's anti-caliphate, in 64/683-4 
(al-Baladjiurl.^aAa^ra/; ivB, 106). In 159/775-6, 
during al-Mahdi's caliphate, 4,000 of the Asawira and 
Sayabidja, together with volunteer troops (muttawwi'-a 
[q.v.] from Basra, were brought together for a naval 
expedition against the coasts of northwestern India; in 
the following year, 760-766-7, this expedition reached 
Gudjarat and the "town of [the river] Narbada", 
probably Broach near the mouth of this river [see 
bharoc] (al-Tabari, iii, 461). After this time, how- 
ever, the Sayabidja fade from mention and were no 
doubt assimilated into the general mixture of popula- 
tion in the high c Abbasid period. 

The etymology of the term Sayabidja, and the 
ultimate origin of this group, were discussed in detail 
by de Goeje, loc. cit., and G. Ferrand in his EI' art. 
s.v., and the validity of their conclusions has not 
subsequently been challenged. Arabic lexicographical 
and other sources describe the Sayabidja as 
mercenaries recruited from Sind, and give as the 
term's singular the form SaybadjJ. De Goeje and Fer- 
rand therefore connected this with Zabadj/Zabag, go- 
ing back to a Sanskritised form Djavaka and a South 

Indian/Dravidian one Shavaka, a term used to 
designate the Indonesian islands of Java and Sumatra 
[see zabad[]. The early Islamic Sayabidja would thus 
be in origin Indonesians who had emigrated to 
western India and who then, in the late Sasanid 
period, found their way to the Persian Gulf shores in 
company with the Zuft. The famed seafaring expertise 
of the Malay-Indonesian peoples would have ensured 
their usefulness to the Middle Eastern powers in such 
matters as the policing of the Gulf and the protection 
of its trade against piracy, etc. 

the article): EI 1 art. s.v. Ch. Pellat, Le milieu basrien, 
40-1, 194, 296; M.J. Morony, Iraq after the Muslim 
conquest, Princeton 1984, 271-2; A. Wink, Al-Hind, 
the making of the Indo-hlamic world, i, Early medieval In- 
dia and the expansion oj Islam 7th-13th centuries, Leiden 
1990, 156-7. (C.E. Bosworth) 

SAYD (a.), a masc. noun and noun of action from 
the root s-y-d which, as in Hebrew, evokes both the 

by earth or sea, which can be eaten as game, 

game, whether caught by hunting or fishing. In all its 
acceptations, the root k-n-s is its exact equivalent. 

The ineluctable need for daily sustenance has led 
mankind, like all other living beings, from the time of 
appearance on earth, to practise both hunting and 
fishing together with the gathering of wild fruits and 
grain, and mankind has accordingly ceaselessly exer- 
cised its ingenuity in finding the best methods here for 
achieving the maximum return for effort. Thus there 
has been a constant striving towards perfection in 
methods of hunting and fishing. 

For hunting game by land, the methods of capture 
(masida, misyada, masyada, pi. masayid) are numerous. 
A passive mode costing the least effort is the setting of 
nets (shabak) and snares with draw-nets (hibala, uhbula, 
pi. habayil); and there is also the covered-over pit-trap 
(hukna, ughwiyya, mughawwat, wadjra, dafina). Then 
comes the method of hunting with a bow iramy, 

aya) or, otherwise, with a cross-bow or blow-pipe 

and, a 

with fire; 

lethods, in the Islamic 
times, one of the most favoured by princes and nobles 
was the chase (iard, mutarada, tirad), on foot or on 
horse-back, with the aid of domesticated or tamed car- 
nivore animals (dari, pi. dawdrt, sayud, pi. suyud), such 
as the gazelle-hound or saluki [q.v.], the cheetah [see 
fahd] and the caracal lynx ( c anak al-ard). These 
precious hunting auxiliaries were launched against the 
gazelle [see ghazal], antelope [see mahat], wild ass 
(himar al-wahsh), ibex (wa% and ostrich [see na c am[. 
According to the region, they sometimes served for 
tracking down wild beasts with furs, such as the pan- 
ther, leopard [see namir] and even the fennec fox [see 
fanak]. Parallel with hunting by the chase, hunting 
by the air, i.e. with raptors [see bayzara], was always 
a favoured pastime for all social classes in the Muslim 
lands, whether high-flying with falcons or low-flying 
with hawks for the capture of small, furry game, such 
as the hare [see arnab] and of that with feathers, such 
as the partridge (hadjal), sand-grouse [see rata], 
bustard (hubdra), wild geese (iwazz), duck (batt) and 
teal (hadhaf). Certain falcons were also trained to tie 
down and blind antelopes and wild assess, thus 
facilitating their capture by the hunting hounds. The 
smallest-sized birds like blackbirds and thrushes were 
taken by means of snares and nooses (sharak, pi. 
ashrak) placed in line on a taut cord. 

For fishing and fish, both in sweet waters and in the 


ing t< 

ilamic law, the hur 

i fisher 



W) v, 

rules regar 


1 slaughter of captured game in orde 
the lawfulness of their consumption; all these rules art- 
set forth in detail by al-Damlrl in his K. al-Hayawanal 
ai-kubra (ed. Cairo 1356/1937, ii, 70-5, s.v. sayd). 
From it, there results that, in order for it to be lawful, 
game should not be killed on the spot and picked up 
dead, so that its throat can be ritually cut. 

Out of all the Arabic authors, lexicographers and 
writers on natural history, there is almost only 
Kushadjim (4th/10th century [q.v.]) who has ex- 
haustively treated all the methods of hunting and 
fishing, in his remarkable K. al-Masdyid wa 'l-matarid 
"Trapping and hunting down" (ed. A. Talas, Cairo 
1954). Four centuries later, Ibn Mangli took up again 
and completed the topic in his substantial K. Uns al- 
mala* bi wahsh al-fala "Relations of the leading men of 
this world with the wild beasts of the endless deserts" 
(tr. F. Vire, Paris 1984). 

Hunting and fishing have inspired most of the great 
poets, such as Aba Nuwas, c Antara, al-Buhturl, Dhu 
'1-Rumma, al-Ghassam, Ibn al-Mu c tazz, Ibn al- 
Rumf, Imru> al-Kays, Labld, al-Nashi>, Ru'ba b. al- 
c Adjdjadj, al-Sanawbarl, al-Shammakh and Zuhayr. 
Kushadjim cited the most noteworthy of these com- 
positions, adding his own ones which had been well- 

Bibliography: See the bibls. to the articles cited 

within the text. (F. Vire) 

SAYDA, the Arabic name of Sidon, the ancient 
commercial city of the Phoenicians. It was cap- 
tured by the Arabs ca. 6/637 by Yazld b. AM Sufyan. 
The Arab geographers mention the fact that, when 
fortified, it was the military port of Damascus. But it 
was not until the period of the Crusades that the city 
achieved any kind of prominence. The Crusaders cor- 
rupted the Arabic name into the form of Sagitta 
(Sagette, Sayette); besieged in 501/1107, it was reliev- 
ed either through the payment of a ransom, or, as the 
Arab authors maintain, as a result of the intervention 
of reinforcements from Damascus. But the city was 
occupied by the Franks in 504/1 1 10 or 505/1 111, after 
a siege lasting 47 days. 

After the departure of the Franks, in 583/1187, 
Salah al-DIn (Saladin) took possession of the city, and 
had the greater part of its fortifications destroyed. 
They were rebuilt by the Franks, who returned in 
625/1228. Recaptured in 647/1249 by the Muslims, 
then in 651/1253 by St. Louis, king of France, who 
fortified it afresh, the city was sacked in 658/1260 by 
the Mongols. The same year, it passed until 691/1291 
into the hands of the Knights Templar. It was then oc- 
cupied definitively by the Muslims. The Druze amir 
Fakhr al-DIn (1003-43/1595-1634) did much that was 
beneficial for the city; the market which he caused to 
be constructed for European traders, the Khan Fran- 
sawi, still exists today. In 1791, the governor Djazzar 
Pasha expelled the French merchants. Then, in 1840, 
Sayda was bombarded by British and Austrian 

The city was incorporated into Greater Lebanon 
under French mandate in 1920, but it was to suffer 
during this period as a result of the monopoly which 
the French accorded to Beirut in respect of maritime 
and commercial traffic. The population, with its Sun- 
ni Muslim majority, vociferously demanded affilia- 
tion with Syria, until the negotiations of March 1936 
between France and its mandated territories; given 
renewed confidence by the arrangements then set in 
place, it participated more actively in national life, 
under the aegis in particular of Riyad al-Sulh, the 
Prime Minister after Independence, who was to be 

assassinated in 1951. It was at the same time respon- 
sive to the calls of'Djamal c Abd al-Nasir advocating 
Arab unity, and it took part in May 1958 in the revolt 
against President Sham c un. 

The population, which had reached a total of 5,000 
inhabitants in the 18th century but was little more 
than 15,000 in 1946, the same number as in 1914, was 
to experience a veritable explosion in 1948, with the 
influx of Palestinian refugees. It quadrupled in 15 
years, reaching a total of 1 17,000 in 1980, half of these 
being Palestinians. With this reinforcement the share 
of the SunnI Muslims in the population of the city ex- 
ceeded 80%. 

From 1970 onward, Sayda and its camps were to 
become a nucleus of Palestinian resistance to Israel. 
The civilian population paid a very high price, endur- 
ing aerial and naval bombardments, even before the 
Israeli invasion of June 1982, which had the object of 
destroying the camps and driving the Palestinians 
towards the north. In February 1985, Israel was oblig- 
ed to withdraw, but its role was taken over by the 
Lebanese forces, the militia of the Christian Right, 
which pounded the camp of c Ayn al-Halwa for a' 
whole month. The Christians of Sayda were then 
forced to abandon the city. 

Palestinian organisations administered the camps, 
and it was not until June 1991 that the Lebanese Ar- 
my took over control of their access points. But while 
the Palestinians have lost their capacity for political 

Sayda has found itself in a position of strength on the 
Lebanese chess-board, with the choice of one of its 
)wn, the multi-millionaire Raflk Hariri, as Prime 

lurbation is composed of num< 


The currei 
very distinct 

The old town occupies the site of ancient Sidon, 
sheltered by a peninsula which encloses the port. Near 
the entry a small island, linked to land by a stone 
bridge, bears the castle of St. Louis. To the south of 
the town, on an artificial hill, stands the Citadel, 
Kal c at al-Mu c izza. Overcrowded and insalubrious, 
the old town has been in an advanced state of 
delapidation since the earthquake of 1956; it accom- 
modates 20% of the population, the poorest, 
Lebanese or Palestinian, and immigrant workers from 
Syria and Egypt. 

A new town, with modern concrete buildings, con- 
taining 46% of the population of Sayda, extends 
beyond the ramparts: the al-Dekerman quarter, 
where public services are concentrated and which is 
the centre of cultural life, and the al-Wastam residen- 
tial quarter, to the north, interspersed with green 

The Palestinian camp of c Ayn al-Halwa, opened in 
1949 by the Red Cross 3 km to the south of the city, 
accounts for 34% of the population of the conurba- 
tion: more than 40,000 people are crammed in there. 

The city has expanded alongside arterial routes, 
particularly towards the east, and it has obliterated the 
orange groves which used to surround it. In fact, it 
has benefited from the fate which dogged Beirut be- 
tween 1975 and 1990, establishing itself as regional 

Sayda's influence extends over the entire region 
between the rivers of Damur and of Litani. Its 
362,000 inhabitants (in 1980), are divided, besides 
Sayda, between three small towns and 140 villages. 

The coastal plain has been devoted since the 1950s 
to market gardening and citrus production, irrigated 
by the al-Kasmiyya canal, which serves 3,700 ha from 
its starting-point at the mouth of the Litani, and by 


hundreds of individual wells and bore-holes. Since the 
1980s plastic greenhouses have proliferated, as a 
result of the capital funds invested by citizens of 
Sayda and of Beirut and by emigres. 

The mountain suffers, however, from lack of water, 
especially to the south. Various irrigation projects 
have been examined, such as that of the Awali, in 
abeyance since 1987, or that of the Djun tunnel, 
which uses a diversion of the LItani for hydro-electric 
production. A pilot project has already succeeded in 
irrigating 1,000 ha in the region of Lebaa, above the 
small Palestinian camp of Miye wa-Miye. 

Cereals and olives are cultivated everywhere, on 
the plains and on the mountain. But this agricultural 
activity has not obviated the need for daily work 
journeys to Sayda or Beirut, nor has it prevented 
large-scale emigration. During the 1970s, many in- 
habitants left for Arabia and the Gulf, following the 
example of the Palestinians from the camps. 

The wholesale market of Sayda takes a third of this 
agricultural production, but currently it is industry 
which is the motive force behind the regional 
economy. Its development has been aided since 1978 
by the influx of Palestinian manpower expelled from 
the south, by the closure of the port of Beirut, which 
led to the diversion of some traffic to the illegal port 
of Sayda, and by the investments of emigres. 

The principal units are the oil refinery, which treats 
Saudi crude, 9 km/5 miles to the south, the power- 
station of Djiye, 10 km/6 miles to the north, the match 
factory of Djiye, the cement-works of SiblTn, and the 
textile factories of c Adlun and of Kfar Djarra. 

The construction sector is active, and light industry 
very diversified: plastics, nylon, cardboard boxes, 
lavatory paper, oils, soap and leather. The local 
sculpted marble, worked by Egyptian craftsmen, is 

This economic development depends on a young 
population, at a rapid rate of growth, and Sayda con- 
tained in 1980 some thirty educational establish- 
ments, public and private. As a result of the policy of 
partition which was powerfully advocated during the 
war, four centres of higher education have been open- 
ed in Sayda : a branch of the Lebanese University in 
1977, of the University of Saint-Joseph in 1978, a 
Makassad Centre of Advanced Studies in 1979 and 
the University of Kfar Falus in 1980. There was a 
total of 3,400 students in 1981-2. 

The city, although no more than 40 km/25 miles 
from Beirut, is thus well equipped to play the role of 
a regional centre. But the potential instability 
resulting from the presence of more than 100,000 
Palestinian refugees continues to cast a dark shadow 
over the future. 

Bibliography: Baladhuri, Futuh, 126; de Goeje, 
BGA, index, s.v.; Ibn al-Athir,, index; 
Yakut, v, 439 ff.; Gildemeister, in ZDPV, viii, 
23 ff.; Baedeker, Paldstina und Syrien; Lortet, LaSyrie 
d'aujourd'hui (1884), 94 ff.; G. Le Strange, Palestine 
under the Moslems, London 1890, index; Abou Saleh 
Salaheddine, La region geographiques de Saida, diss. 
Paris 1987; D. Chevallier, La societe du Mont-Liban 
a I'epoque de la revolution industrielle en Europe, Paris 
1971; Hasan Hammoud, L'amenagement regional au 
Liban. Etude d'un cas type: le caza de Saida, diss. Paris 
1992; Elizabeth Picard, Liban, etat de discorde, Paris 
1988; Nadine Picaudou, La dechirure libanaise, Paris 
1989; Rosemary Sayigh, Too many enemies. The 
Palestinian experience in Lebanon, London 1994. 

(M. Lavergne) 
al-SAYDANA (al-Saydala) (a.) is pharma- 
cology, in the meaning of pharmacognosy. The 

druggist is called al-saydaldni or al-saydandni. Al-BTrunl 
defines him as follows: "He is someone who occupies 
himself with gathering medicaments according to 
their most commendable sorts and with selecting their 
best kinds, both simple ones and those which have 
been prepared according to the most excellent com- 
positions, which have definitely been determined for 
that purpose by medical authors". Elsewhere he says 
"al-saydana therefore is the knowledge of simple drugs 
according to their selected sorts, kinds and forms, as 
well as the knowledge of the mixture of medicaments 
composed in conformity with their written prescrip- 
tions or on the basis of what the trustworthy and 
righteous researcher strives for. The highest rank, 
however, is held by the knowledge of the effects of the 
simple medicaments and their specific qualities" [see 
adwiya]. Besides, saydana indicates the druggist's ac- 
tual store of drugs, and also (with or without a 
preceding kilab) the handbook of drugs, the phar- 
macopeia. Al-saydaldni is practically synonymous with 
al-'attdr [q.v. ; almost everything said there is also valid 
for al-saydaldni] . Since healing powers are ascribed 
to many perfumes, both terms indicate also the 
merchant of spices and aromas [see afawih, in 

The classical definition of pharmacology is found in 
al-BTrunT's [q.v.] highly important Preface to his un- 
finished K. al-Saydana ]fi 'l-tibb], written in his old age. 
He classifies al-saydana as a sub-discipline in the field 
of medicine. According to him, it is the first of the 
stages of the medical art and for many it counts only 
as the latter's preliminary stage because it is a tool for 
practising medicine, not a part of it. As far as the 
word saydana is concerned, al-BIrunl first refers to the 
well-known fact that the Arabic sad corresponds with 
Indo-Iranian am. He approvingly quotes Hamza al- 
Isfahani [q.v.], who is said to have explained saydanani 
as an Arabisation of candandni "merchant of san- 
dalwood" . Sandalwood is not a medicinal plant par ex- 
cellence, but one may assume — al-BIruni continues — 
that the Persians, when looking for sandalwood, came 
in contact with the Indians and called their merchant 
of perfume candandni; the Arabs would then have 
taken over this term, and Arabised it because they did 
not know a name for it. And since sandalwood was not 
counted among their perfumes, and since they were 
hardly able to distinguish between a perfume mer- 
chant and a drug merchant, they identified the two 
words. For the peculiar scent of sandalwood, see san- 
dal. There exist, therefore, in this case the same tran- 
sitions of meaning as those found in the case of '■attar. 
The second consonant of clsandandni, the nun, would 
then have been miswritten or misread into yd*. The 
fact that the word saydalalsaydalani, as far as is known 
to the writer of these lines, is completely unknown in 
the western Muslim world, may be explained from its 
Indian origin; in the West, the corresponding terms 
are [ c ilm] al-adwiya al-mufrada or al-murakkaba, in 
quotations often abbreviated to al-adwiya [see adwiya] 
or ['Urn] al-'uturl al-'attdr (see above). 

As an oriental synonym of al-saydaldni, al-BIrunl 
also mentions al-ddri, with which the Arabs in the old 
days indicated the perfume merchant, because the 
ships from India brought their goods to the port of 
Darin, lying in the area of al-Bahrayn. Al-BIrunl 
substantiates the meaning of the word with examples 
taken from ancient Arab poetry. 

A general theory of pharmacology is given by al- 
MadjusI [see c ali b. al- c abbas al-madjusi] in his 
Kdmil al-sina'a al-tibbiyya, Bulak 1294/1877, ii, 84-100. 
The section is comprehensively analysed by M. 
Ullmann, Islamic medicine, Edinburgh 1978, 104-6. 


Bibliography: The most important study on 
saydana is that by M. Meyerhof, Das Vorwort zur 
Drogenkunde des Biruni, in Quellen und Studien zur 
Geschichte der Naturwissenschaflen und da Medizin, iii, 
Berlin 1932, fasc. 3, esp. Arabic text 3-8, tr. 25-37. 
Meyerhof took the Arabic text from the unique 
manuscript Bursa, Kursunluoglu (composed 
678/1278). A Turkish translation of the Preface was 
made by §erefeddin Yaltkaya, Birunlu Eba Reyhan 
Kitabussaydele fittibb mukaddimesi, in Istanbul Univer- 
sitesi Tib Tarihi Enstitiisii, x, Istanbul 1937. An edi- 
tion of the complete Arabic text, with an English 
tr. , has been given by Muhammad Sa c id, Biruni. K. 
al-saydana fi 'l-tibb. Al-Biruni's Book on Pharmacy and 
Materia medica, Karachi 1973. There exist a Russian 
tr. by U.I. Karimov, Abu Raykhan, Farmakognoziya v 
medicine (K. al-Saydanafi 'l-tibb), issledovaniye, perevod, 
primecaniya i ukazateli, and a Persian tr. by c Abbas 
Zaryab, Biruni, Kitdb al-Saydana fi 'l-tibb, Tehran 
1369/1949. _ (A. Dietrich) 

al-SAYDAWI, Shams al-DIn Muhammad al- 
Dimashkl, outstanding musician and writer on 
music in Syria in the second half of the 9th/15th cen- 
tury. Born in Sayda at an unknown date, he later liv- 
ed in Damascus where he died on 16 Dhu '1-Ka c da 
911/10 April 1506. 

Al-Saydawf composed an extensive didactic urdjuza 
of nearly 250 verses on the musical modes (anghdm) of 
the Syrian tradition, entitled K. al-ln'dm (or An'am)fi 
ma'rifat al-angham. In addition to the twelve main 
modes (four labelled asl and eight jar'-) and the six so- 
called awaz modes he describes another seven secon- 
dary modes called buhur (sing. bahr). To illustrate their 
melodic development, al-SaydawI uses stave systems 
of seven coloured lines representing the degrees of one 
octave. Within the stave he marks the principal notes, 
the direction of melodic motion and some other details 
of performance by using coloured symbols and ab- 
breviations derived from musical terms such as 
ma^khadh and rakz (initial and final note), su'ud and 
hubut (ascending and descending motion), slow mo- 
tion (bi 'l-tartib; "step by step"), quick motion 
{sur'atf"), sustaining of notes (madd), and places of 
"jumping" towards higher notes (matdfira, sing, mat- 
jara). The whole range of three octaves is represented 
by abdjad letters in the colours of the basic octave. As 
far as we know at present, al-Saydawi's musical nota- 
tion is unique in Arabic (and also Persian and 
Turkish) music literature. It has, however, cognates 
in the stave systems of the Graeco-Latin Musica en- 
chiriades (9th century) and in those given by Vincenzo 
Galilei (1581) and Athanasius Kircher (1650). The 
" considerable 

'little . 

excitement when it reached Paris in If 
Diderot's Encyclopedic {Planches, vii, 3-4) and in 
d'Herbelot's Bibliothique orientate (ii, 758), and it was 
partly translated, around 1780, by Pigeon de Saint- 
Paterne on behalf of de La Borde. 

Besides the K. al-ln'am, al-Saydawf left some 
religious poetry, mostly muwa shsh ahat. including a so- 
called ndtik, a didactic poem in which each verse is 
sung to another mode. The ndtik was popular in the 
16th and 17th centuries and still appears in the song 
text collection (ms. Berlin, or.oct. 1027) of the Syrian 
litterateur Ibn al-Khal (d. 1117/1705). 

Bibliography (in addition to the titles mentioned 
above): GhazzI, al-Kawakib al-sdHra, Beirut 1945, i, 
79; J.-B. de La Borde, Essai sur la musique ancienne 
et moderne, Paris 1780, i, 180-90; H.G. Farmer, 
Historical facts for the Arabian musical influence, London 
1930, 323-6; A. Shiloah, The theory of music in Arabic 
writings, Munich 1979, 83-6 (corr. name and 

dating); A. Shiloah and A. Berthier, A propos d'un 
' 'petit livre arabe de musique' ' , in Revue de musicologie, 
lxxi (1985), 164-77. (E. Neubauer) 

SAYF IBN DHI YAZAN, Sirat, an Arabic 

nature, inspired in a very remote fashion by the life 
of the eponymous individual [q.v.) 

Known in numerous manuscript versions, of which 
the earliest dates from the 1 lth/1 7th century, the story 
was probably composed in Mamluk Egypt between 
the 9th/15th and 10th/16th centuries; the identity of 
the hero's principal antagonist, Sayf (Sayfa) Ar c ad, 
emperor of Ethiopia from 1344 to 1372, rules out an 
earlier date. As for the attribution of the story to Abu 
'1-Ma c all (Muhammad b. c Abd al-Baki, d. 991/1583- 
4), mentioned in numerous versions, this should be 
treated with caution, even though this author may 
have played a role in the composition of the work. The 

what might be expected, the old Yemenite historico- 
legendary tradition is in fact not much in evidence. 
On the other hand, a number of themes are percepti- 
ble which belong to the Midrashic tradition centred 
especially on the figure of Moses (A. Chraibi, Dialecti- 
que de la sur determination dans le Sirat Sayf ibn DI Yazan, 
in Arabica, forthcoming), also themes and motifs 
relating to international folklore, as well as foundation 
legends which seem typically Egyptian; also to be ob- 
served are convergences with shorter stories belonging 
to the genre of the Thousand and One Nights, in par- 
ticular the Tale of Adjib and Gharib. It should be stress- 
ed that, until the present time, a meticulous nar- 
ratological study of the novel in its entirety has yet to 
be undertaken; the same applies to a systematic com- 
parison of the different manuscript versions, which 
might give a clearer impression of the process of com- 
position of the story. The present article will be based 
essentially on the analysis of the printed version, 
which is characterised by the abundance of fantastic 
and marvellous themes, often with a parodic in- 

Unlike other major Arabic popular romances, such 
as the Sirat 'Antara, the Sirat Dhdt al-Himma, the Sirat 
Baybars, even up to a point the chronicle of the Banu 
Hilal, which are all located in a definite and 
recognisable historical framework in spite of 
numerous anachronisms, the Sirat Sayf is situated from 
the outset in the mythical primordial universe, where 
men and djinns associate together on familiar terms, 
where sorcerers, wizards and enchanters engage in 
dogged combat, competing for power or for the 
mastery of natural forces. This universe is steeped in 
impiety and the worship of false gods; only a few sages 
and a few anchorites observe and cherish the "Law of 
Abraham", which it will be the hero's task to pro- 
pagate among the djinn of Mount Kaf. From the point 

jor parts, a "Yemenite cycle" and an "Egyptian 
cycle", between which are interspersed various ad- 
vantures following the classical pattern of the 
"quest". The Yemenite cycle, after a prologue evok- 
ing the conquests of Dhu Yazan, the father of the pro- 
tagonist, and the foundation by his vizier Yathrib of 
what is to become Medina, destined to be the refuge 
of the future Prophet, follows a familiar theme. Sayf, 
abandoned in the desert on the orders of his mother, 
the sorceress Kamariyya, who intends to take power 
into her own hands, is rescued by the king Afrah. Fall- 
ing in love with the latter's daughter, the beautiful 
princess Shama, he is obliged to undergo various 
tests, in the course of which he establishes himself as 
the greatest warrior of his time, discovers the secret of 


his birth, converts to monotheism and regains the 
throne of his fathers, having obtained the hand of his 
beloved (this marriage being only the first of a long 
series, the "Law of Abraham" being considered to 
allow unlimited polygamy). As well as to the courage 
of the hero, his success is owed to the help which he 
receives from a small group of loyal companions, 
among whom there appear numerous female charac- 
ters, in particular the djinniyya c Akisa, daughter of the 
White King and foster-sister of the protagonist, as 
well as the sage and magician c Akila. Among the 
enemies of Sayf, besides Kamariyya, who is executed 
after committing innumerable treacheries, figure the 
powerful Abyssinian emperor Sayf Ar c ad and the two 
wizards Sakardls and Sakardiyun. The latter 
recognise in Sayf the hero whose destiny it is to imple- 
ment the curse laid by Noah upon Ham and his 
descendents, dooming them to fall under the domina- 
tion of the sons of Shem; however, the conflict seems 
to be primarily religious, with the Yemenite 
"Muslims according to the religion of Abraham" op- 
posing the Abyssinian "worshippers of stars" and, 
more generally, pagans of all allegiances. 

The first of the two "quests" serving as a link to the 
second cycle takes up the familiar folkloric theme of 
the search for a supernatural spouse who has disap- 
peared (J.E. Bencheikh, CI. Bremond and A. Miquel, 
Mille et un conies <k la nuit, Paris 1991, 193-233). The 
second is constituted by an amalgam of various 
adventures, linked by a somewhat tenuous theme: to 
fulfil a promise made to the djinn c Ayrud, who has 
fallen in love with c Akisa, Sayf sets out in search of the 
wedding jewellery of the Queen of Sheba, preserved 
among Solomon's treasure. A secondary "mini- 
cycle", possibly of Syrian origin and evoking the 
adventures of Dummar, elder son of Sayf, leads into 
the second section of the novel: returning from his 
quest, Sayf finds that his capital has been destroyed by 
Sayf Ar c ad during his absence. On the advice of 
c Akila, he decides to lead his people into Egypt, which 
is still a desert and inhospitable land: the Nile, halted 
by a powerful spell, is blocked at the level of the 
Cataracts. Thanks to numerous supernatural helpers, 
Sayf succeeds in "liberating" the waters of the river. 
The remainder of the story recounts the foundation 
and the organisation of the new kingdom (numerous 
foundation legends), then the hero's final revenge on 
Sayf Ar c ad as well as on Sakardls and Sakardyun; in 
the printed version, this is pursued through various 
episodes which are clearly later additions. 

Despite some tedious passages and occasional clum- 
siness of style, the romance, which is of great interest 
for the study of Arabic narrative traditions, is not 
without appeal on account of its unrestrained ex- 
ploitation, not devoid of irony, of marvellous and fan- 
tastic themes, which form the basis of a genuine 
literary work. 

Bibliography: Strut al-Malik Sayf ibn Dhl Yazan 
fdris al-Yaman, 17 vols., Cairo 1881-5 (numerous 
reprints); F. Khurshid, Strut Sayf, Cairo 1982 
(abridged version of the "Yemenite cycle", preced- 
ed by an Introduction); E.W. Lane, Manners and 
customs of the modern Egyptians, London 1836, ch. 
xxii; R. Paret, Strut Saifibn Dhijazan, Ein Arabischer 
Volksroman, Hanover 1924; idem, EP s.v.; J. 
Chelhod, La geste du rot Sayf, in RHR, clxxi/1 
(January-March 1967), 181-205; Th. Mankus, Sayf 
b. Dhl Yazan, bayn al-haktka wa 'l-ustura, Baghdad 
1980; H.T. Norris, Sayf b. Di Yazan and the book of 
the history of the Nile, in Quaderni di Studi Arabi, vii 
(1989), 125-51. (J. -P. Guillaume) 

SAYF b. C UMAR, a compiler of historical nar- 

rations on early Islamic history. Virtually 
nothing is known about Sayf or his life, except that he 
lived in Kufa and probably belonged to the Usayyid 
clan, part of the c Amr branch of the tribe of Tamlm. 
Various sources, however, attach him — probably 
erroneously — to other tribal groups; accounts stating 
that he was of the Asad tribe stem from misreading his 
nisba "al-Usayyidi" as "al-Asadl", but he is also said 
to have belonged to the tribes of Dabba and Dubay c a, 
or to the Baradjim, a group of five Tamlm clans (not, 
however, including Usayyid). According to the 
8th/14th-century scholar al-Dhahabl, Sayf died in the 
time of al-Rashfd (r. 170-193/786-809), but this may 
be merely al-Dhahabfs guess, deduced from Sayf s 
position in various isnads. 

Sayf's importance rests on the fact that his Kitab al- 
futuh al-kabir wa 'l-ridda was chosen by the famous 
historian al-Tabarl (d. 3 10/923) as his main source for 
the ridda and the early Islamic conquests. Sayf is also 
one of the few sources of information about the struc- 
ture of the early Muslim armies, early Muslim 
government, and settlement in the garrison towns. 
The reliability of Sayf's narrations has long been con- 
tested, however, beginning already with the 
mediaeval hadtth specialists and their biographers, 
who noted the suspect character of his hadtths; some 
accused him of zandaka, others simply noted that he 
put fabricated accounts (mawdu'-at) in the mouths of 
trustworthy transmitters. Many modern scholars, 
after examination of both the content and the isnads of 
Sayfs accounts, have expressed similar doubts. 
Already de Goeje and Wellhausen pointed out the im- 
plausible chronology of Sayfs accounts when com- 
pared to those of Ibn Ishak, al-Wakidl, etc. The fan- 
ciful quality of many of Sayfs narrations, which are 
frequently fitted out with gratifying but suspect detail, 
and the tribal chauvinism of many accounts, in which 

frequency in battlefield heroics, have also provoked 
critical comment. Sayfs isnads have also been seen as 
problematic; his informants frequently cannot be trac- 
ed in the available literature on traditionists. This has 
led at least one modern scholar, Murtada al- c Askan, 
to condemn Sayf as a subversive who invented his ac- 
counts and the names of his informants with the intent 
of spreading confusion and doubt in the bosom of the 

On the other hand, Landau -Tasseron has pointed 
out that Sayfs accounts frequently do not contradict 
other accounts, but merely provide different informa- 
tion, and has argued that we cannot expect many of 
Sayfs informants to appear in the traditionist 
literature since most were not specialists in hadtth. 
Furthermore, Sayf retails not only pro-Tamim tradi- 
tions, but also accounts that highlight other tribes, 
suggesting that he may have collected various tribal 
traditions from Kufa. Many of his accounts display 
syntactic and lexical peculiarities that may be reflec- 
tions of archaic tribal dialects. Close inspection of 
Sayfs accounts conveyed from Abu c Uthman Yazid 
b. Asid al-GhassanT, an informant identifiable in no 
other source, suggests that they are not fabricated 
wholesale by Sayf, but represent selections from two 
different written sources, one compiled by Abu 
c Uthman himself, the other by an intermediary in- 
formant who drew on Abu c Uthman, each work 
having a different topical focus. 

Comparison of the accounts from Sayf found in al- 
Tabari with those found in Ibn c Asakir's Ta'rtkh 
madinat Dimashk suggests that Sayfs original compila- 
tion on the ridda and conquests may have consisted of 
a long, sustained narrative with occasional isnads 


(rather than the short pieces, each prefaced by an 
isnad, that are found in al-Tabari" s and Ibn c Asakir's 
selections). Comparison also reveals that the original 
work of Sayf featured frequent poems, most of which 
have been edited out of al-Tabari' s extracts. 

Sayf's information on institutional matters has been 
virtually indispensable to historians. Yet uncertainty 
persists about the reliability of his narrations of the 
ridda and conquests, which raises the question why al- 
Tabari, who is generally praised for his sober 
historical judgment, favoured Sayfs narrations of 
these events over others that seem less dubious. Hum- 
phreys attributes al-Taban's choice to the eirenic and 
moralising qualities of Sayfs interpretation of Islamic 
history, which smoothed over the conflicts that 
historically divided the Companions of the Prophet. A 
definitive study of the historiographical complexities 
of all Sayfs traditions remains an important 

Bibliography: Dhahabi. Mizdn al-iHiddl, Cairo 
1325, ii, 255-6, no. 3637; Ibn Hadjar, Tahdhib al- 
tahdhib, Haydarabad, 1325-7, iv, 295-6, no. 506; 
Ibn Abl Hatim, al-Djarh wa 'l-ta^dil, Haydarabad 
1952-3, ii, 278; Fihrist, 94; F. Sezgin, GAS, i, 311; 
M.J. de Goeje, Memoire sur la conquete de la Syrie, 
2 Leiden 1900; J. Wellhausen, Prolegomena zur dlteslen 
Geschichte des hlams, Berlin 1899 (=Skizzen und 
Vorarbeken, 6); Djawad C A1I, Mawdrid taMkh al- 
Tabari, in Madfallal al-madjma c al-Hlmi al-Hrdki, i 
(1950), 179-82, ii (1951), 163-6, viii (1961), 427-9; 
Murtada al- c AskarI, Khamsin wa-mPa sahdbi mukh- 
talak, i, Baghdad 1969; A. Noth, Der Charakler grofier 
Sammlungen von Nachnchten, in 1st., xlvii (1971), 168- 
99; M. Hinds, Sayf b. 'Umar's sources on Arabia, in 
Studies in the history of Arabia. 1. Sources for the history 
of Arabia, part 2, Riyad 1979, 3-16; A. A. Duri, The 

1983, index; Ella Landau-Tasseron, The participa- 
tion of TayyP in the Ridda, in JSAI, v (1984), 53-71; 
idem, Sayf ibn '■Umar in medieval and modern scholar- 
ship, in 1st., Ixvii (1990), 1-26; R.S. Humphreys, 
Translator's foreword, in The history of al-Tabari, xv. 
The crisis of the early Caliphate. The reign of c Uthmdn, 
Albany 1990, pp. xv-xvii; idem, The odd couple: al- 
Tabari and Sayf b. 'Urnar, forthcoming in L.I. Con- 
rad (ed.), History and historiography in early Islamic 
times: studies and perspectives, Princeton 1995; F.M. 
Donner, Translator's foreword, in The history of al- 
Tabari. x. The conquest of Arabia, Albany 1993, pp. 
xvi-xx; A. Noth and L.I. Conrad, The early Arabic 
historical tradition. A source-critical study, Princeton 
1993, index. (F.M. Donner) 

SAYF al-DAWLA, Aeu '1-Hasan c AlI b. Abi >1- 
Haydja' c Abd Allah b. Hamdan b. Hamdun b. al- 
Harith Sayf al-Dawla al-Taghlibl (17 Dhu '1-Hidjdja 
303-24 Safar 356/22 June 916-9 February 967), amir 

Mayyafarikin and of western Djazira (Diyar Bakr and 
Diyar Mudar), from 333/945 until his death. From his 
time until the present day, he has personified the Arab 
chivalrous ideal in its most tragic aspect. A peerless 
warrior, magnanimous vanquisher of rebellious 
Syrian tribes, he led with audacity, and for a long 
time with success, the dphdd against the Byzantine 
enemy. A prince of great wealth, he spent large sums 
of money on the ransom of Muslim prisoners. An 
enlightened poet and philologist, he maintained a 
literary court more brilliant than that of any other 
Islamic sovereign. A man of Shi<T leanings, he suf- 
fered terrible ordeals towards the end of his life, defeat 
at the hands of the Byzantines, the brutal killing of 
members of his family, debilitating illness and the 

;. This in- 

treachery of his most trusted li 
dividual has inspired an abundan 
as well as historical, of which a fi 
given here. Al-Dhahabi appends to his name the 
following qualificatives: maksad al-wufud, ka'-bat al-djud, 
faris al-islam, hdmil liwap al-djihdd (Sirat aHam al-nubala?, 
cd. Arna 3 ut and al-Bushi, Beirut 1983, xvi, 187, 
henceforward abbreviated to Sim; Abu '1-Fadl 
Bayhakf, TaMkh, Arabic tr., Cairo 1956, 408, a long 
eulogy by al-Tha c alibi; Ibn Khallikan, Wafaydt al- 
a'ydn, ed. Ihsan 'Abbas, iii, 401). 
The poetic circle 

Among the men of letters, the majority of them 
from c Irak or from Djazira, whom he invited to his 
court or who praised him in order to benefit from his 
generosity, certain names are worth mentioning: the 
khatib of the djihad Ibn Nubata al-FarikT, the philoso- 
pher Abu Nasr Muhammad b. Tarkhan al-Farabl, 
the poets Abu Firas al-Hamdani (various accounts of 
his captivity, A.-M. Edde, Description de la Syrie du nord 
d'lbn Shadddd, Damascus 1984, 288, henceforward 
abbreviated to Description), Abu '1-Tayyib al-Mutan- 
nabbl (on the last-named, see Ibn al- c Adim, Bughyat 
al-tdlib, ed. Zakkar, henceforward abbreviated to 
Bughya, 639-86), Abu '1-Kasim Nasr b. KhaHd al- 
Shayzami. the amir's foil, Abu 'l- c Abbas Ahmad b. 
Muhammad al-Nami al-MissisT, Abu '1-Faradj c Abd 
al-Wahid al-Babbagha J , Abu '1-Faradj Muhammad 
al-Wa 5 wa>, Abu Bakr al-Sanawbari, Abu '1-NadjIb 
Sadad b. Ibrahim al-Djazarl al-Zahir, Abu '1-Hasan 
al-Sarl b. Ahmad b. al-Sarl al-Kindi al-Raffa 3 al- 
Mawsili, the two brothers al-Khalidi Abu Bakr 
Muhammad and Abu c Uthman Sa c id, sons of 
Hashim b. Wa c ula al- c Abdi al-MawsilT, al-Tarsusi al- 
Nadjranl, Abu Muhammad al-Mawsill, Abu c Abd 
Allah al-Ghamr b. Abi '1-Ghamr al-KhalI c al-Shami, 
Ibn Batta, al-Wassaf, the Kilabi" poet al-A c sar b. 
Muharish, the Sa«y poet AbO c Abd Allah Ahmad b. 
al-Husayn al-Aftasi al-Sakran, the composer of 
despatches and poet Abu '1-Fadl b. Salim al- c Utaridi 
al-Manbidji, the kdtib Abu c Ali Ahmad b. Nasr al- 
Baziyar, the astrologer Abu c Abd Allah al-Baghdadi, 
Abu '1-Hasan al-Karkhl, Aba c Abd Allah al-Basrl, 
AbO Muhammad al-TinmsI, Abu '1-Husayn c Ali b. 
Muhammad al-Shimshatl, Ibn Sadaka al-Mawsill al- 
Nahwl, Aba C A17 al-Hasan b. Ahmad al-FasawT al- 
Farisi al-Nahwi al-LughawI, Abu '1-Tayyib < A txj al- 
Wahid b. C A1T al-HalabT al-Lughawi, Aba C A1T 
Muhammad b. al-Hasan al-Lughawi al-Hatimi, the 
Hanaff kadi Aba Sa c Id al-Hasan b. c Abd Allah b. al- 
Marzuban al-Sayraff al-Nahwi, the 'drnil of Antioch 
AbO c Abd Allah al-Husayn b. Muhammad b. al-Sakr 
al-Ma c lathawT al-Mawsill, the kdtib al-Husayn b. C A1I 
al-Maghribl and his son C A1I, the Sufi Abu '1-Kasim 
Ahmad b. Humaydan al-Rummani, Ibn al- 
Khashshab and the Mu c tazili mutakallim Aba Bakr 
Ahmad b. c Ubayd Allah Ibn Bint Hamid, who con- 
ducted a theological controversy regarding the Kur'an 
with Aba c Abd Allah al-Husayn Ibn Khalawayh, 
tutor to the sons of Sayf al-Dawla, in the latter's 
presence. The accounts concerning them demonstrate 
that all topics were apparently dealt with, religious — 
hadith, kaldm, the Kur 3 anic readings— as well as 
linguistic, historical, philosophical, astronomical, 
astrological or purely literary. Poetical contests were 
particularly frequent. The index of Bughya, 5212, 
under c Alt b. c Abd Allah Sayf al-Dawla, makes it 
possible to trace the biography of the majority of the 
individuals who adorned the literary soirees of the 
amir at Aleppo or at Mayyafarikin (see also M. Ca- 
nard, Histoire de la dynastic des Wamdanides, hence- 
forward abbr. to Wamd., 18, 33; Brockelmann; Sez- 


gin; and EP, under the writers mentioned). The 
accounts concerning al-San, Bughya, 4202-10, 4403, 
describe in a number of instances the assemblage of 
poets waiting in the vestibule, al-dihliz, to be received 
at the court of Sayf al-Dawla, in the hope of being 
awarded a few hundred dirhams. The influence of this 
court extended far afield. The poet Sari, having been 
obliged to leave Sayf al-Dawla on account of the 
jealousy of the al-Khalidi brothers, lived miserably in 
Baghdad, hounded by their hatred; hence his jubila- 
tion, recounted by Ibn al- c Adim, when he received 
much later from a wazir a gift of 3,000 dirhams. He 
felt able to show himself in the street, wearing fine 
clothing and a turban five cubits in length, followed 
by magnificent slaves, then he died of happiness. In 
all the Islamic capitals of the time, the parasitism of 
the courtier-intellectuals was justified by the prop- 
aganda which their works disseminated to the advan- 
tage of the local prince. The poets, in particular, com- 
peted fiercely for the attainment of glory and wealth, 
bearing witness to the glory and the wealth of the 
prince. At Aleppo, the major official poets such as al- 
Mutanabbi maintained in their turn a court of secon- 
dary poets around them (Bughya, 4514-15), which did 
not prevent the latter from abandoning Sayf al-Dawla 
for Kafur in 346/957, the year of the great conspiracy. 

On the other members of the family, see ham- 
danids. The nisba al-Taghlibi links this family to an 
Arab tribe, living in Djazira in the c Abbasid period. 
Ibn Hazm is the only one who describes the Banu 
Hamdan as mawali of the Banu Asad. The family 
always maintained very close ties with the Kurdish 
tribes, and was possibly itself of Kurdish origin 
(Wamd., 287-8, 305-6). 
Early military career of the young Hamddnid 

While in Baghdad, the political authority of the 
c Abbasid caliph was subordinated to that of the amir 
al-umara > Ibn Ra°ik, in Diyar Rab^a and at Mawsil 
the governor was al-Hasan b. c Abd Allah b. Hamdan, 
later renowned under the lakab of Nasir al-Dawla. In 
324/936, the latter promised his brother, C A1I b. c Abd 
Allah, the future Sayf al-Dawla, aged 21 lunar years, 
the gift of Diyar Bakr in exchange for his aid in con- 
fronting the Daylami C A1I b. Dja c far, governor of 
Mayyafarikfn, who was in revolt. C A1I b. c Abd Allah 
succeeded in preventing C A1I b. Dja'far from receiving 
the assistance of the feudal chiefs of Armenia and ob- 
tained from his brother the amdn appointing him 
governor (Wamd. 478). Then he intervened at Diyar 
Mudar, in the region of Sarudj, against the Kaysi 
tribes of Numayr and Kushayr which were creating 
disorder, and in 325/937 he took official possession of 
this province. In 328/939-40, C A1I b. c Abd Allah sum- 
moned the princes of the region, Armenians and 
Georgians, Christian or Muslim Arabs, to a meeting 
on the shore of Lake Van. He obtained their submis- 
sion and their assistance in assuring the security of the 
Djazira by means of a network of fortresses, along the 
routes linking Persia, c Ir5k and Syria with Byzantine 
Anatolia, the Black Sea, the Caucasus and the 

In 330/942, on behalf of his brother who had acced- 
ed in his turn to the post of amir al-umarp and had re- 
ceived the lakab of Nasir al-Dawla, C A1I b. c Abd Allah 
victoriously confronted Abu '1-Husayn al-Baridi near 
al-Mada'in, and received on this occasion the lakab of 
Sayf al-Dawla. For the first time, lakabs in the form of 
-dawla were awarded to persons who did not occupy 
the post of wazir, head of the caliphal civil administra- 
tion, but were soldiers, one of whom exercised no 
supreme authority (Wamd., 426, Hilal al-Sabi 3 , 
RusCm dar al-khilafa, ed. Mikhail c Awwad, 127-8). 

Sayf al-Dawla was then apparently appointed to the 
post of wali al-harb wa 'l-saldt at Wasit, which led to his 
confrontation with Ahmad b. 'All al-Kufi, formerly 
the wazir in practice of Ibn Ra 3 ik and fiscal ad- 
ministrator of the same province, in regard to the duty 
to be levied on a barge. The attitude of the two pro- 
tagonists reveals the depth of the corruption which 
was then rife in 'Irak, involving financiers, private 
sector traders and the holders of civil or military 
power. In 331/943, Sayf al-Dawla encountered op- 
position on the part of Turkish officers and soldiers 
under his command over the issue of pay; this opposi- 
tion was led by Tuzun. Nasir al-Dawla was powerless 
to rescue him, being himself in difficulties in 
Baghdad, the Samanid amir of Transoxiana and the 
Ikhshld of Egypt, who coveted Mawsil (Wamd., 445), 
having refused to help him. Sayf al-Dawla reached 
Baghdad and then joined his brother, who had taken 
refuge at Mawsil after the nomination of Tuzun as 

The seizure of Syria 

In 328/939-40, Ibn Ra 3 ik had detached northern 
Syria from the control which the I khsh ld of Egypt had 
exercised there since 324/935-6, but in 329/941 he left 
Damascus with the aim of regaining supreme power 
in Baghdad and in Radjab 330/ April 942, he was 
assassinated at Mawsil. Nasir al-Dawla wanted to 
return to northern Syria, which the Ikhshld was intent 
on re-occupying, to the direct control of the c Iraki 
caliphate. His troops easily took possession of the 
valley of the Balikh, from Harran to Rakka, but cer- 
tain of his officers sided with the I khsh ld. and the 
Hamanids who had remained loyal were confronted, 
in the region of Khabur, by the governor of Rahba, 
c Adl al-Bakdjami, and by the Banu Numayr. With the 
surreptitious support of the I khsh ld. and supplied 
with Turkish and Daylami contingents, c Adl marched 
on Naslbfn, where he took possession of the treasure 
of Sayf al-Dawla. Abu c Abd Allah al-Husayn b. Sa c Id 
b. Hamdan, the brother of the poet Abu Firas, cap- 
tured c Adl, who was executed in Baghdad in Sha c ban 
331/May 943. Nasir al-Dawla then entrusted this 
cousin, Abu c Abd Allah b. Sa c id, with the task of tak- 
ing control of northern Syria, of Diyar Mudar and of 
the frontier posts. Having been obliged to take Rakka 
by assault, Abu c Abd Allah occupied Aleppo without 
difficulty in Radjab 332/943-4. 

Sayf al-Dawla was quite prepared at this time to 
take control of Syria himself; it was an idea born of 
resentment when, having returned to Nasibin, he 
found himself under-employed and badly paid. He 
campaigned against the Banu Numayr in the region 
of al-DSliya but, betrayed by his Arab troops, he was 
defeated in the winter of 332/943-4 on a number of oc- 
casions by Tuzun, and he joined the caliph al-Muttaki 
at Rakka, hoping to have himself recognised in his 
turn as amir al-umara\ but he encountered the caliph's 
hostility in this matter and had his rival, Muhammad 
b. Inal al-Turdjuman, assassinated (various accounts 
concerning this topic, Wamd., 497). Having agreed to 
recognise Tuzun's sovereignty over c Irak, from Sinn 
to Basra, his brother Nasir al-Dawla had officially re- 
nounced his claim to the title amir al-umara 1 in ex- 
change for recognition of his authority over Djazira 
and northern Syria, from Mawsil to the limits of 
Syrian territory. 

During the summer of 332/944, the Ikhshld arrived 
at Aleppo, which was hastily abandoned by al-Husayn 
b. Sa c id. In Muharram 333/September 944, the Ikh- 
shld presented himself in his turn at Rakka before the 
caliph al-Muttaki, to whom he offered his services; 
obtaining no definite response, he returned to Egypt. 


Nasir al-Dawla was alleged then to have said to Sayf 
al-Dawla, with whom he was reconciled: "Syria lies 
before you, there is no one in this land who can pre- 
vent you taking it." His cousin Husayn b. al-Sa c Id 
having renounced his Syrian ambitions in his favour, 
Sayf al-Dawla decided to take control of Aleppo and 

Sayf al-Dawla made his entrance into Aleppo in 
Rabf I 333/October 944, in the company of Abu '1- 
Fath 'Uthman b. Sa c id b. al- c Abbas b. al-Walld al- 
Kilabl, a tribal chieftain whose brother Abu 'l- c Abbas 
Ahmad was then governor of the city and who had 
himself previously held the post on behalf of the Ikh- 
shld. Having regained the capital of northern Syria, 
the Hamdanid and the Kilabi toured the region from 
the crossing of the Euphrates riding in the same 
c ammariyya, with the aim of observing at first-hand the 
condition of the villages. 

The I khsh id was not readily prepared to accept the 
loss of northern Syria and of the frontier towns facing 
the Byzantines. He opposed Sayf al-Dawla when the 
latter revealed his aspiration to bring Hims into his 
domain. The Hamdanid routed an Ikhshidid army 
commanded by Kafur at Rastan and occupied Hims. 
From c Ayn al-Djarr, last stage in the Bika c before the 
crossing of the Anti-Lebanon, he sent to the in- 
habitants of Damascus a letter which was read on the 
minbar of the Great Mosque (Th. Bianquis, Les derniers 
gouverneurs ikhshidides a Damas, in BEO, xiii [1970], 
186). The Ikhshid left Egypt at the head of an army 
in Ramadan 333/April 945. He proposed to accept 
Sayf al-Dawla's authority over northern Syria, the 
djunds of Hims and of Kinnasrln, and over the frontier 
sites. Sayf refused, but was obliged to leave Damascus 
as a result of the hostility of the populace, and joined 
battle with the Ikhshid in Shawwal 333/May-June 
945. Defeated by means of a strategic ruse, Sayf al- 
Dawla withdrew to Rakka. The Ikhshid army ravaged 
the neighbourhood of Aleppo and maltreated the 
population. In Rabf I 334/October 946, a treaty on 
the basis of the I khsh ld's former proposals was 
negotiated by the intermediary of the Husaynid shanf 
Abu Muhammad al-Hasan b. Tahir b. Yahya al- 
Nassaba {Wamd., 584, details of the negotiations in 
Bughya, 2408-12, and Ibn Sa c Id al-AndalusI, al- 
Mughrib fi hula al-Maghrib, Cairo 1953, 194). A parti- 
tion of Syria was agreed: to the south of the new fron- 
tier, Tarabulus, Ba'labakk, Labwa {Bughya, 371) and 
Damascus remained attached to Ikhshidid Syria; to 
the north of this line, c Arka, Djusiya and Hims form- 
ed part of the principality of Aleppo. The Egyptian 
leader also committed himself to paying an annual 
tribute to compensate for the renunciation of claims to 
Damascus on the part of Sayf al-Dawla, who was 
obliged to marry the daughter of c Ubayd Allah b. 
Tughdj, the Ikhshld's brother. The signing of the 
ceremony gave the caliph al-Muti* the opportunity to 
ratify the amirate of northern Syria and the lakab of 
Sayf al-Dawla by means of the sending of ceremonial 
attire which arrived in Aleppo in the spring of 335/946 
(al- c AzimI, TaMkh Halab, ed. Za c rur, 291, hencefor- 
ward abbr. to al- c Az!mi). 

In Dhu '1-Hidjdja 335/July 947, the Ikhshid died in 
Damascus. Sayf al-Dawla immediately marched on 
this city where he tried in vain to confiscate the minor 
estates of the Ghuta (Ibn al- c Ad!m, Zubda, ed. Saml 
Dahhan, IFEAD, Damascus 1951, i, 116-17, 
henceforward abbr. to Zubda). He left for Palestine, 
but defeated by Kafur, the black eunuch leading the 
armies of Unudjur b. al-Ikhshld. in Djumada I 335/ 
December 946, the Hamdanid withdrew to the region 
of Damascus, where he was unobtusively reunited 

with his mother, then retreated towards Hims, follow- 
ing the eastern route by way of Kara, a difficult route 
in winter. In the spring of 335/947, having rallied 
troups among the Arab tribes of c Ukayl, Numayr, 
Kalb and Kilab, Sayf al-Dawla advanced on 
Damascus and was once more defeated by the Ikhshid 
army which entered Aleppo in Dhu '1-Hidjdja 
335/July 947. In the autumn of 336/947, Sayf al- 
Dawla re-occupied Aleppo, this time definitively. 

Pursuing the I khsh ld's policy of appeasement, 
Kafur negotiated with Sayf al-Dawla. The previous 
treaty was ratified but the Egyptians retained 
Damascus definitively and no longer paid tribute to 
the Hamdanids (topographical details of the Ham- 
danid possessions in northern Syria, Zubda, i, 164-5, 
on the occasion of the signing of the armistice between 
Karghuya and the Byzantines in 359/968). This fron- 
tier between northern Syria, inclined towards c Irak, 
the Djazlra and Anatolia, and southern Syria, closely 
linked to Egypt and Arabia, was 



the province in 658/1260. 

During the remainder of his life, Sayf al-Dawla had 
no further confrontations with the Ikhshlds. The 
Egyptian power, threatened at this time by the 
Fatimids installed in Ifrlkiya, gave up its claims to 
northern Syria, according to the judicious principle 
whereby abandoning a peripheral province was less 
costly than maintaining an army on a war-footing and 
less dangerous for the authority of the amir. Master of 
a large territory, the traditional glacis of Egypt, com- 
prising the Mediterranean coast from Tarabulus al- 
Gharb to Tarabulus al-Sham, and the Red Sea as far 
as Yemen, the son of the I khsh id retained the greater 
part of his father's former power and wealth, leaving 
to the Hamdanid the costly defence of the frontier 
facing resurgent Byzantium. 

Sayf al-Dawla was henceforward the master of 
northern Syria, of the western and eastern frontier 
posts of the Diyar Mudar and Diyar Bakr. 
Mayyafarikln, at the junction between these two 
western provinces of the Djazlra. was his second 
capital. The embellishment of the city and the rein- 
forcement of its defences were to be continued under 
the Kurdish Marwanid dynasty [q.v. ] which made it 
its capital, in the 5th/llth century. Sayf al-Dawla 
built at Mayyafarikln municipal and defensive con- 
structions, on a considerably larger scale than his 
earlier buildings in Aleppo (on the Palace of Halba 
constructed by Sayf al-Dawla outside Aleppo and 
traversed by the Kuwayk, Bughya, 349-50, 4489, 
Canard, Sayf al Daula, recueil de textes, henceforward 
abbr. to Recueil, 204-5, Wamd., 642-4). 

For Diyar Mudar and Diyar Bakr, Sayf al-Dawla 
remained theoretically the delegate of his brother, the 
prince of Djazlra. resident at Mawsil. For northern 
Syria and the frontier regions, a less clearly defined 
chain of vassality existed since the 'Abbasid caliphate, 
despite the sending of an envoy to Sayf al-Dawla in 
335/946, seems from 332/943-4 onward to have 
recognised only one provincial authority, Nasir al- 
Dawla, receiving "in appanage" all the territories to 
the north and north-west of c Irak as far as the 
Mediterranean and the Byzantine frontier. On the 
other hand, functioning in a Kurdish and DaylamI 
environment, the Hamdanid family had embraced the 
rule of primogeniture, and Sayf al-Dawla always 
showed great respect towards his elder brother, 
although the latter' s real power was much inferior to 
his own (well known "unbooting" episode, Zubda, i, 
128-9, Bughya, 2433-7, Wamd., 621-2). Furthermore, 
he appointed numerous Hamdanids, including sons of 


Nasir al-Dawla, to posts of authority in his states 
(Wamd., 595-7, supplies the list). 

After 336/947 he visited Djazira on several occa- 
sions. The first time, in 338/950 he travelled by way 
of Rakka to Mayyafarikln, where in turn his mother, 
Nu'ra and his son, Abu '1-Haydja 3 c Abd_ Allah had 
recently died, and returned to Aleppo via Amid. This 
expedition, at the head of 7,000 troops, was an oppor- 
tunity to display his wealth and the power of his army. 

Since 332/943, the Kurd Daysam had resided in 
Aleppo as a guest of Sayf al-Dawla. In 344/955 he left 
for Adharbaydjan, where he established the khulba on 
behalf of the Hamdanid but, defeated by Marzuban 
and handed over to the latter by the Armenian prince 
of Vaspurakan, Derenik/Ibn al-Diranl, he was put to 
death. Sayf al-Dawla lost his influence in Armenia un- 
til 354/965 when, following the revolt of his ghulam 
Nadja, he inherited the Armenian places which the 
latter had appropriated from Abu '1-VVard/Apelbart II 
b. Abi Salim, in the region of Lake Van. The fate of 
these places when Abu Taghlib, the nephew of Sayf 
al-Dawla, took possession of Diyar Bakr, is unknown. 
Suppression of the tribes 

Sayf al-Dawla was obliged to assert his authority in 
northern Syria by confronting the Arab tribes, whose 
aggressiveness had been increased by a very positive 
demographic situation. At the time of his arrival, the 
region of Hims had been controlled since the 
Umayyad period by the Yemeni tribes of Tayyi 3 and 
Kalb, semi-sedentarised. To the north, the Kaysi 
tribes, c Ukayl, Numayr, Kilab, Ka c b and Kushayr, 
still nomadic, held the plain between the Orontes and 
the Euphrates and beyond the river, in Diyar Mudar, 
threatening longer-established Arab communities. 
Taghlib had been repulsed to the north and east of the 
Djazira (attempt at devising a map of the tribes in 
Bianquis, Rahba et le Diyar Mudar, ... in BEO, xli-xlii 
[1989-90, publ. 1993), 23-53). The region of Ma'arrat 
al-Nu c man was the fief of the Yemeni Tanukh, 
formerly converted to Christianity and entirely seden- 
tary. The coastal positions between Tarabulus and al- 
Ladhikiyya were inhabited by the Yemeni Bahra 3 and 
by Kurdish groups. 

In 336-7/947-8, after a difficult siege, Sayf al-Dawla 
took possession of the fortress of Barzuya, held by a 
Kurdish brigand who controlled the lower valley of 
the Orontes and the route joining Aleppo to al- 
Ladhikiyya. He campaigned throughout the littoral 
range and on the coast, as well as on the slopes of the 
valley of the Orontes where his primary objectives 
were to defend the interests of the Bahra^ and to pro- 
tect the Kalb, who were his allies against the Kilab. At 
the end of the winter of 337/949, a Karmatl agitator 
by the name of Ibn Hirrat al-Ramad al-Kharidji, who 
had claimed the titles of al-Hadi and of Sahib al-khal al- 
Mubarka^, incited the Kalb and the Tayyi' of central 
Syria to revolt, and imprisoned Abu Wa'il, the gover- 
nor of Hims (Wamd. , 603). Sayf al-Dawla crushed the 
rebels, killed the Karmatl and freed Abu Wa'il. The 
following year, another revolt associated with the 
Karmatis took place in the region of Damascus; this 
time, the rebels were c Ukaylids, i.e. KaysTs. 

The KaysT tribes regularly caused instability in 
northern Syria between 338/950 and 343/954, pro- 
voking limited campaigns of repression on the part of 
the Hamdanids (Wamd., 602-18). The Bedouin refus- 
ed to return to the arrangement whereby they were 
kept away from the cultivated zones, where they 
wanted to pasture their cattle and extort ransoms from 
the villagers. The revolt of winter and of spring 
344/955 erupted to the south-east of Aleppo, in the 
region of the lake of Djabbul and of Kinnasrin. This 

was not a tribal movement but a protest against social 
conditions since it united Bedouin, Yemenis, Kalb 
and Tayyp, and Kaysls, c Amir b. Sa c sa c a, Ka c b b. 
Rabi c a, c Adjlan b. c Abd Allah, Kilab, Numayr and 
Kushayr, thus prefiguring the great uprisings against 
the Fatimids of the following century. The repression 
imposed by Sayf al-Dawla in the course of June of the 
same year is known to the smallest detail, as a result 
of the analysis by Canard of the poems of al- 
Mutanabbf and of Abu Firas, as well as of commen- 
taries on their work. This was a desert policing opera- 
tion perfectly planned and rigorously executed. It 
could have ended with the total extermination, 
through warfare and thirst, of all the tribes, women 
and children included, between Salamiyya, Tadmur 
and the Euphrates, if Sayf al-Dawla had not been in- 
fluenced by his feelings of solidarity and his sense of 
Arab honour. Later, the heavy armies of the 
Fatimids, composed of Berber horsemen and Turkish 
ghulams, had neither the same ability to win battles 
with minimal loss of life nor the same sensitivity in 
avoiding unnecessary civilian casualties. 

Sayf al-Dawla never again had to face Bedouin 
revolts. The Numayr had been expelled to the north 
of the Euphrates, to the Diyar Mudar around Harran, 
where they were to form a principality in the following 
century, the Kalb migrated towards Tadmur and into 
the plain between Hims and Dimashk, the Tayyi 3 
took refuge in the Djawlan and on the plateaux of 
Transjordania. The small Kays! tribes having been 
either eliminated or banished to the north of the 
Euphrates, the Banu Kilab were left as the only 
powerful formation of northern Syria. Wisely, they 
became reconciled to the Hamdanid and sent armed 
contingents to serve under him. They pursued a 
policy of assistance to the official power and of 
peaceful penetration of the urban milieu which was to 

cipality of the Banu Mirdas [q.v.]. Once security had 
been restored, Sayf al-Dawla entrusted to relatives or 
to close associates the delegation of local powers based 
on fortified sites (al- c Azimi, 192). The central ad- 
ministration at Aleppo and at Mayyafarikin was 
limited, being confined to fiscal issues and the conduct 

The i 

against Byzantium 

Sayf al-Dawla was motivated throughout his life by 
the desire to be aghazi, a knight of the djihdd, who con- 
fronted the Byzantines in more than forty battles. 
Numerous texts, prose or poetry (Bughya, 4682; 
Recueil, passim) reveal the extent to which the period 
was marked by a well-founded fear of a Byzantine of- 
fensive against the Muslim lands of Cilicia and of 
Djazira and by a militaro-religious brand of piety 
which induced professional soldiers and young 
civilians, from all over the Muslim East, to risk their 
lives on the frontier facing Byzantine troops. The 
significant role of Tarsus in the ideology of religious 
asceticism and of the military efficacy of the frontier 
djihad against the Byzantines, is developed in Bughya, 
180-1, which mentions Sayf al-Dawla (see also C.E. 
Bosworth, The city of Tarsus and the Arab-Byzantine fron- 
tiers in early and middle c Abbasid times, in Oriens, xxxiii 
[1992], 268-86). 

A detailed description of the frontier military opera- 
tions of Sayf al-Dawla having no place here, mention 
of the most important dates will suffice. In fact, 
Canard has compiled a complete account of combats 
between Hamdanids and Byzantines, based on the de- 
tailed analysis of a very large corpus of Arabic, Greek, 

veil a 

on his 

l regarding historical geography {Wamd., 


715-862, including for the period of Sayf al-Dawla, 
741-828). Few new texts have been published since 
then, although recent works dealing with the strategy, 
tactics, arms and military training of the Byzantines 
deserve mention (G. Dagron, Guerilla, places-fortes et 
villages ouverts a lafrontiere orientale de Byzance vers 950, in 
Castrum, iii [Madrid-Rome 1984]. In the Arab do- 
main, no comparable recent study exists to this 
writer's knowledge. 

Unlike the Byzantine basileis who, supported by the 
substantial resources of an extensive empire were 
equipped to pursue, after 314/926, a project of recon- 
quest of eastern Anatolia, of Cilicia and of northern 
Syria, Sayf al-Dawla, amir of a reduced principality, 
had no strategy other than a policy of defence of the 
frontier, reinforcement of the towns which were 
threatened, and reconstruction and repopulation of 
the latter after their destruction by the Byzantines. 
This defensive activity was interspersed with raiding 
expeditions into the interior of Byzantine territory: 
"they threw themselves against al-Safsaf and Wad! 
Sabur, they burned the towns, took the children into 
captivity, they slaughtered the parents, they took for- 
tresses by assault; this was a terrible and glorious ex- 
pedition" (Bughya, 4293; Recueil, 87). Such attacks 
provided booty and facilitated the exchange of 
prisoners, but no mention is made of a predetermined 
plan for the methodical conquest of 


.ught tt 


i, preparation for battle, placing of 
mbushes, the use of a variegated army made up of 
iverse ethnic groups, Daylamis, no doubt the most 
s, Turks, Kurds, Arabs, negroes, and their 
well as the cost of recruitment, 
e and operations of the army, the cost of 
the damage caused by occupations and revolts, and 
the relationship between this cost and the various 
fiscal measures which financed the war. The corpus of 
sources should make it possible to go even further 
than Canard went in this domain. 

The Byzantines continued with their effort at 
breaking through on the Armenian front. Even before 
becoming amir of Aleppo, in 324/936 the future Sayf 
al-Dawla had led an expedition against the Byzantines 
and in support of Sumaysat [q.v.\, which was to fall 
soon afterwards. His first victory dates from the 
autumn of 326/938, confronting the Domesticus John 
Corcuas, who commanded a large army. In 328/939- 
40, he conducted operations in Armenia in order to 
secure the allegiance of the Armeno-Arab amirs, and 
he acquired on this occasion renown as a champion of 
the djihdd in the Muslim world. Once installed in 
Aleppo, Sayf al-Dawla engaged in limited operations 
against the Byzantines in the winter of 333/945, fol- 
lowed in autumn 335/946 by an exchange of 
prisoners, but real hostilities did not begin until 
336/947, when the treaty with Kafur acknowledged 
Sayf al-Dawla's domination of northern Syria and 
responsibility for the frontier with Byzantium. 
The triumphant years 

In the thirty years which followed, Sayf al-Dawla 
was frequently at war, and the theatre of operations 
was situated at this time in the southern sector of the 
frontier, between Diyar Bakr and the Medi 
principally in Cilicia; although master of the 
region of northern Syria, he never engaged ir 
warfare. It was the Egyptian fit 

t the 1 

t this 

: (al- 

c AzImI, 298-9). The wars conducted after his installa- 
tion at Aleppo and Mayyafarikin may be divided into 
two periods. From 336/947 to 346/957, Sayf al-Dawla 
enjoyed overall success, then, from 350/961 to his 

death in 356/967, he faced with fortitude a series of 
grave defeats and saw even Aleppo briefly occupied by 
the Byzantines, then threatened again. He had to deal 
first with the Domesticus Bardas Phocas, already an 
old man, then with his three sons Nicephorus, Leo 
and Constantine, formidable strategists. 

In 336-8/947-50 the Byzantines had taken Mar'ash, 
then Kallkala, threatening Tarsus. Sayf al-Dawla 
organised the response. After some success, his cam- 
paign ended in the autumn of 339/950, near Buhayrat 
al-Hadath, with total defeat, the ghazwat al-musiba, the 
calamitous expedition, from which he personally 
escaped by a miracle. To strengthen his base of opera- 
tions, in 341/952 Sayf al-Dawla rebuilt numerous 
Cilician strongholds, including Mar c ash, and was 
thus able to repulse a Byzantine offensive. Constan- 
tine Porphyogenitus sent a delegation to Aleppo with 
the offer of a treaty, but in vain. During the summer 
of 342/953, near Hadath, commanding a reduced 
cavalry force, Sayf al-Dawla crushed the army of the 
Domesticus, who was wounded and one of whose sons 
died in captivity at Aleppo. With the same eloquence 
with which they had bemoaned his defeat, the poets 
immortalised the victory of Sayf al-Dawla, who once 
again refused to negotiate a treaty (on the complex 
relationship between Byzantium and Aleppo, see 
Wamd., 778-9). Another victory over the ponderous 
army of the Domesticus in the autumn of 343/954 al- 
lowed Sayf al-Dawla to complete the reconstruction of 
Hadath, which successfully withstood a Byzantine ex- 
pedition during the summer of 344/955. The old 
Domesticus Bardas Phocas, a miser and a poor tacti- 
cian, was then replaced by his son, the future basileus 
Nicephorus II Phocas, Patricius and Strategus of the 
Anatolica, Domesticus of the Schola, assisted by his 
younger brother Leo and by the Armenian John 
Tzimisces. In the spring of 345/956, when Tzimisces 
tried to intercept him on his return, laden with spoils, 
from an expedition towards Hisn Ziyad and Tall 
Bitrlk, Sayf al-Dawla won another victory, having 
crossed a stream on material furnished by the 
engineer corps of the army. In the autumn, he made 
his way to Cilicia to support Tarsus, threatened by the 
advancing fleet of the Theme of Cibyrrheotes. 
The successful Byzantine response under Nicephorus Phocas 
However, his fortunes changed, and Sayf al-Dawla 
was unable to avenge the capture and destruction by 
the Domesticus of the fortress of Hadath during the 
summer of 346/957 because he discovered a con- 
spiracy of his senior officers in Aleppo to deliver him 
to the Byzantines in the course of a campaign; he 
ordered the execution of 180 ghulams and more than 
200 others had hands, feet and tongues amputated 
and torn out (Zubda, i, 127-8)._The summer and 
autumn of 347/958 saw near Amid a victory of 
Tzimisces over Nadja, the Circassian ghulam who was 
a favourite of Sayf al-Dawla, and the fall of Sumaysat 
and of Ra'ban, places defended by Sayf al-Dawla in 
person. With the reorganisation of the Byzantine ar- 
my, the Muslim front was entrusted to Leo Phocas, 
who in the winter of 348/959 inflicted a bloody defeat 
on Sayf al-Dawla and occupied numerous fortresses 
between Cilicia and Diyar Bakr. Nicephorus Phocas 
having commited himself from the summer of 960 to 
the spring of 961 to the occupation of Crete, Sayf al- 
Dawla seized the opportunity to launch a raid in com- 
pany with the inhabitants of Tarsus. On his return, 
Leo Phocas inflicted on him a terrible defeat in 
Ramadan 349/November 960. The Arab amirs of the 
frontier regions sided with Byzantium, and Ibn 
Zayyat made the khutba at Tarsus in the name of the 
caliph, not mentioning Sayf al-Dawla, whose power 


was challenged even at Aleppo. From the autumn of 
350/961 to the spring of 963, Nicephorus Phocas, 
returning victorious from Crete, led the offensive into 
Muslim territory. c Ayn Zarba and some fifty villages 
were taken in the winter of 350-1/961-2. Nicephorus 
massacred part of the civilian population and 
destroyed the fruit trees, inaugurating the policy of 
terror which he conducted against the Arabs until his 
death. During the summer of 351/962, Sayf al-Dawla 
managed partially to reconstruct c Ayn Zarba. In Dh u 
'1-Ka c da 351/962, Nicephorus Phocas marched on 
Aleppo; Sayf al-Dawla did not become aware of the 
Byzantine advance until they were approaching 
c Azaz. He had few troops with him, his army being 
far away under the command of Nadja, hence was 
defeated under the walls of Aleppo. Nicephorus forced 
his way into the city, which was pillaged and 
devastated. He did not succeed in taking the citadel 
(or rather the height defended by a DaylamT garrison 
on which the citadel was later to be built), since 
towards the end of the year 351/962 he was obliged to 
leave the city and return hastily to Constantinople, 
where he was proclaimed emperor. He had taken 
away with him 10,000 young captives, and Sayf al- 
Dawla, in order to repopulate and rebuild his capital, 
installed there the inhabitants who had fled Kinnasrln 
and used the subsidies sent to him by his sister. After 
a few minor successes during the summer of 352/963, 
at the end of the same year the Muslims suffered a 
crushing defeat at the hands of the new Domesticus, 
Tzimisces, at the Hill of Blood near Adana. The year 
353/964 saw little military activity. Sayf al-Dawla, a 
sick man, was confined to his own territory by revolts, 
and famine raged in Cilicia, making the land un- 
suitable for large-scale operations. The Byzantine of- 
fensive was resumed during the winter of 353-4/964-5 
and, again on account of the famine, did not end until 
the following summer with the fall of Missisa/Mop- 
suesta and of Tarsus. Despite a rescue attempt by the 
Egyptian fleet which arrived too late, the latter city ex- 
perienced in Sha'ban 354/ August 965 an honourable 
capitulation; the inhabitants were allowed to leave for 
Muslim territory with their property. Cilicia was 
henceforward methodically re-Christianised. An ex- 
change of prisoners was negotiated in 355/966; the 
Muslims were obliged to ransom a further 3,000 men 
at a cost of 200,000 gold pieces. Byzantine operations 
were now directed against Syria, and Nicephorus 
Phocas marched on Manbidj, and then on Aleppo and 
on Antioch before returning to Byzantine territory at 
the end of 355/966. Two months later, Sayf al-Dawla 
died in Aleppo. 
Military reverses, domestic revolts and physical weakness 

Since 346/957, Sayf al-Dawla had not had to con- 
tend with internal revolt. In 350/961, the rebellion of 
Muhammad b. al-Husayn Ibn al-Zayyat, amir of Tar- 
sus, in favour of the Ikhshldid Unudjur, was quickly 
suppressed {Bughya, 1490, 3419-21). In 351/962, the 
brief occupation of Aleppo by the Byzantines cost him 
his military prestige, at a time when he had recently 
been stricken by hemiplegia. From this time onwards, 
handicapped by the after-effects of this attack, ag- 
gravated by increasingly serious intestinal and 
urinary disorders, Sayf al-Dawla remained bed- 
ridden. He was transported on a litter, but he retained 
his reason and continued to direct the policies of the 
principality and, when necessary, to conduct warfare. 
However, his political position was fragile. The 
leading ghuldms were impregnated with the notion 
dominant in the Muslim East, according to which 
political power was legitimised solely by success in 
battle. In 352/963, Hibat Allah b. Nasir al-Dawla, 

governor of Harran, whence he had been expelled by 
a populace resentful of his fiscal greed, killed at Alep- 
po Abu '1-Husayn Ibn Danha, a Christian secretary 
of Sayf al-Dawla who had the latter's absolute trust, 
then attempted to incite his administrative charge to 
rebel in favour of Nasir al-Dawla (Miskawayh and 
Ibn al-Athlr, cited in Recueil, 247-52). At the approach 
of Nadja, whom Sayf al-Dawla had sent to oppose 
him, Hibat Allah sought refuge with his father. On his 


orted c 


i of z 

million dirhams from the inhabitants of Harr; 
set out, in his turn, to attack Mayyafarikin with the 
intention of placing Diyar Bakr under the control of 
MuSzz al-Dawla, the Buyid amir of Baghdad. The 
wife of Sayf al-Dawla defended the city successfully, 
Nadja took possession of towns in Armenia, around 
Lake Van, and in the autumn of 353/964 he was 
obliged to abandon a second siege of Mayyafarikin, a 
revolt having broken out in his Armenian dominions. 
Having returned to Mayyafarikin, Sayf al-Dawla set 
out for Armenia, where personally and without 
fighting he obtained the spectacular submission of his 
favourite lieutenant (moving scene in Zubda, i, 145); 
Nadja was killed at the end of winter 354/965 in 
Mayyafarikin, no doubt at the orders of Sayf al- 
Dawla's wife whom he had publicly insulted. He was 
buried in the Hamdanid mausoleum. 

Sayf al-Dawla had suffered a personal blow in 
352/953 with the death of his sister, Khawla Sitt al- 
Nas, a woman of considerable political skill. She be- 
queathed to him 500,000 dinars which he is said to 
have devoted to the ransom of prisoners (Sira, 16, 
188). In 354/965 his sons Abu '1-Makarim and Abu '1- 
Barakat died. He was affected then by depression but 
continued to defend his power. During the last phase 
of his life, he resided mainly in Mayyafarikin, en- 
trusting the defence of the western frontier to his 
senior ghuldms. Kinnasrln was abandoned by its in- 
habitants, and commerce, the principal source of 
revenue of this cross-roads region, was hampered. 
The civilian populations of the region of Aleppo ac- 
cepted their fiscal burdens with increasing reluctance, 
while the armies of the amir proved themselves in- 
capable of protecting them from the Byzantine 
enemy. In 354/964, to prevent the acquisition of iron 
by the Karmatls of al-Ahsa', who intended to re-arm 
themselves in preparation for eventual action in Syria 
and Egypt, he did not hesitate to dismantle the gates 
of the town of Rakka, which were constructed of 
armour-plated wood. An alliance with the Karmatls, 
opposed to the Buyid of Baghdad, contributed at the 
same time to the maintenance of stability on the 
Syrian plain among the Arab tribes, Yemeni or 
Kaysi, over whom they had retained strong influence. 
The Karmatls were furthermore hostile to Kafur. The 
latter, based on Fustat, continued to take an interest 
in Syria, intent on protecting Damascus from any 
Hamdanid interference. 

In 354/965 there erupted the far more dangerous 
revolt of the former prefect of the littoral, the KarmatI 
Marwan al- c Ukayli, who took possession of Hims, 
defeated a Hamdanid army commanded by the ghulam 
Badr and reached Aleppo but was wounded and died 
shortly afterwards (different versions, Wamd., 649, 
and Zubda, i, 148). In the autumn of 354/965 (dated 
incorrectly in H\md., 651), Abu '1-Hasan Rashlk b. 
c Abd Allah al-NasIml, former governor of Tarsus, a 
town henceforward in the possession of Nicephorus 
Phocas, impelled by a tax-farmer, al-Hasan b. al- 
AhwazI, took possession of Antioch and assembled an 
army which enabled him for three months to besiege 
Aleppo, defended by Karghuya and Bishara. (On this 


revolt, see Bughya, 3656-8, 4592-3; Descr., 240-1.) 
After partially occupying the lower town, RashTk was 
killed in battle. He was replaced as leader of the 
rebellion by a Daylaml, Dizbar, who defeated 
Karghuya and entered Aleppo but then abandoned 
the town with the intention of taking control of all 
northern Syria. Sayf al-Dawla was then carrying out 
an exchange of prisoners with the Byzantines at 
Sumaysat, and he saw the Muslims, ransomed by 
him, joining the ranks of the Daylami rebel. At the be- 
ginning of the summer of 355/966, Sayf al-Dawla had 
himself carried in a litter to Aleppo, where he spent 
the night and then set out to crush the troops of Diz- 
bar, in which he was helped by the defection to his 
side of the Banu Kilab. Reprisals were severe, 
although the new governor appointed to Antioch by 
Sayf al-Dawla, TakI al-DIn Muhammad b. Musa, 
joined the ranks of the Byzantines with the public 
funds of the city. The attack by Nicephorus Phocas on 
northern Syria forced Sayf al-Dawla to take refuge at 
Shayzar, until Safar 356/February 967, the date of his 
death in Aleppo to which he had just returned (ac- 
cording to al- c Azimi, 303, he died at MayyafarikTn). 
The sharif Abu c Abd Allah al-Aksasi prayed over his 
body, pronouncing five takbirs in the Shi'T fashion 
(Bughya, 4517). 

The ShM sympathies of Sayf al-Dawla are demon- 
strated by the al-Dakka mashhad which he had built on 
the Djabal Djushan alongside the Christian 
monastery of Mara Marutha (Bughya, 412, 2726). 
After the murderous invasion of Aleppo by the Byzan- 
tines in 351/962, he invited sharifs from Harran or 
from Kum to take up residence in Aleppo. The 
replacement of Sunnism by Twelver Sh/ism as the of- 
ficial and popular religion of Aleppo may thus be 
dated from the mid-4th/10th century. Some of their 
descendants, notably the Banu Ibn Abi '1-Djinn 
(Bug/v/ia, 2415-16; Bianquis, Damas et la Syrie sous la 
domination fdtimide, IFEAD, Damascus, index, 
henceforward abbr. to Damas et la Syrie) played an im- 
portant role in the Fatimid period. The delimitation 
of the judicial madhhab applied at Aleppo in the 
4th-5th/10th-llth centuries whether Hanafism or 
Imamism, seems difficult to identify precisely (oral 
indication of Anne-Marie Edde). On the relationship 
between Sayf al-Dawla and Isma c Ilism and the diffu- 
sion of Nusayrism in Syria in the 4th/10th century, 
the role of al-Husayn b. Hamdan al-KhasIbl, of 
Muhammad b. C A1I al-Djilli and of Surur al- 
Tabarani, see Canard, Wamd., 634, Halm, art. 
nusayri, and Bianquis, Damas el la Syrie, 375-6. 

His embalmed body was transported to 
Mayyafarikin to be laid to rest in a mausoleum, 
beside his mother and his sister. A brick containing 

operations against the Byzantines was placed under 
his cheek; this was almost the interment of a shahid. 
He left only two living offspring: a daughter, Sitt al- 
Nas, and a son, Abu '1-Ma c ali Sharif, to whom he had 
had the bay'a made before his death (Sira, 16, 188) and 
who as prince of Aleppo bore the lakab of Sa c d 

Sayf al-Dawla was served in the office of wazir by 
Abu Ishak Muhammad b. Ibrahim al-Kararlti, 
formerly in the service of the chief amir Ibn Ra'ik, 
then of the caliph al-Muttaki; subsequently, he 
dismissed him, appointing in his place Abu c Abd 
Allah Muhammad b. Sulayman b. Fahd. Finally, 
Abu '1-Husayn 'All b. al-Husayn al-Maghribl [see al- 
maghribI, banu] was an efficient and loyal wazir, no 
doubt working with his father, al-Husayn b. C A1I. At 
the time of Sayf al-Dawla's installation at Aleppo, the 

kadi was Abu Tahir Ahmad b. Muhammad b. Mathil. 
Finding him insufficiently obsequious, Sayf al-Dawla 
replaced him with a friend of the poet Abu Firas, Abu 
Husayn c Ali b. c Abd Allah (or b. c Abd al-Malik) al- 
Rakki, whom Sayf al-Dawla later accused of having 
induced him to commit iniquitous acts. He was killed 
by the Byzantines in 349/960, and his son Abu '1- 
Haytham was to be wazir of a descendant of Sayf al- 
Dawla. The same year, Sayf al-Dawla awarded the 
charge to Ibn Mathil. Abu Dja c far Ahmad b. Ishak al- 
Hanaff was to succeed the latter, and was exercising 
this function at the time of the death of the amir of 
Aleppo (Zubda, i, 112, 113, 131, 152). 
Confiscation of land by the Hamddnids 

On the basis of the information given by the 
chroniclers as well as by the geographers, it should be 
possible to reconsider the balance-sheet of the Ham- 
danid dynasty. The wars between Sayf al-Dawla and 
the Ikhshid had led to the destruction of the olive- 
groves and orchards surrounding Aleppo. The agri- 
cultural landscape was permanently altered (Bughya, i, 
415-16, Bianquis, Le pouvoir politique a Alep au VIXP 
Steele, in REMMM [1992], no. 62, 49-59). Further- 
more, this prince as well as his brother, Nasir al- 
Dawla and later, his nephew al-Ghadanfar. princes of 
Mawsil, had a policy of monopolising fertile agri- 
cultural lands and appropriating extensive domains, 
with the aim of devoting them to monoculture, in par- 
ticular the growing of cereals, a decidedly profitable 
enterprise in view of the demographic growth of 
Baghdad. Combining these territorial confiscations 
with oppressive fiscal policies, they became the 
wealthiest amirs of the Islamic world (Ibn Hawkal, 
Suratal-ard, 111/12, 177-89/174-86, 200-24/207-30, in 
particular 211-14/205-8; Bughya, 1, 59, 10, 4237, 
4593). They thus acquired lasting glory by showering 
with precious gifts their kinsmen and the poets who 
eulogised them, but they seriously destabilised 
agriculture and crafts, commercial exchanges and the 
equilibrium between towns and countryside, on the 
plains and plateaux of DjazTra and of northern Syria. 
Ibn Hawkal, followed by Ibn al- c Adim, dates from the 
confiscations of Sayf al-Dawla the ruin of the city and 
the commerce of Balis. By destroying orchards and 
peri-urban market gardens, by enfeebling the once 
vibrant polyculture and by depopulating the seden- 
tarised steppe terrain of the frontiers, the Hamdanids 
contributed to the erosion of the deforested land and 
to the seizure by semi-nomadic tribes of the agri- 
cultural lands of these regions in the 5th/llth century 
(Wamd., 397, 436). 

Bibliography: Given in the text; for Arab as well 
as Christian sources, see M. Canard, Sayf al-dawla, 
recueil de textes, Algiers 1934; Histoire de la dynastie des 
Wamdanides de Diazira et de Syrie, i, Algiers 1951 , in- 
dex; Arabica, xviii (1971), 279-319; to be sup- 
plemented by the bibliography at the end of c Izz al- 
Din Ibn Shaddad, al-AHik al-khatira fi dhikr umard> 
al-Shdm wa l-DJazira, Description de la Syrie du Nord, tr. 
Anne-Marie Edde, IFEAD, Damascus 1984; edited 
text, in BEO, xxxii-xxxiii (1980-1), 265-402. The 
principal texts which have appeared since 1951 had 
been used in manuscript form by M. Canard; these 
are al- c AzImi, Ta\ikh Halab, ed. Ibrahim Za c rur 
(unreliable), Damascus 1984, and especially Ibn al- 
c AdTm, Zubdat al-halab min ta\ikh Halab, i, ed. Sami 
Dahhan, IFEAD, Damascus 1951, and Bughyat al- 
talab fi ta\ikh Halab, i-x and index, ed. Suhayl Zak- 
kar, Damascus 1408/1988. For source material re- 
garding Sayf al-Dawla in the Bughya, Ibn al- c Adim 
used a great many works, including a Sirat Sayf al- 
Dawla, or Akhbar Sayf al-Dawla, a work of Abu '1- 


Hasan C A1I b. al-Husayn al-Daylaml, see Bug/pa, 
2531. The sources used by Ibn al- c AdIm for the 
4th/10th century have not been listed to this writer's 
knowledge, see Edde, Sources arabes des XIV et XIII' 
siecles d'apres le dictionnaire biographique d'Ibn al- c Adim, 
in Itineraires d 'Orient. Hommages a Claude Cahen, 
Bures-sur-Yvette 1994, 293-307. A few volumes of 
Ibn c Asakir, Ta?ri~kh madinat Dimashk, have also ap- 
peared, but the rate of publication by the Arabic 
Academy of Damascus remains regrettably slow. 
See also Ibn Abl Usaybi c a, c Uytln al-anbi^ fi 
labaqdt al-atibba' ', ed. Tahhan, Cairo. The editions 
of Sibt Ibn al-DjawzI, Mir'dt al-zamdn, for the years 
350-450 and of DhahabI, Ta\ikh al-isldm, for the 
4th/10th century, have not been available for con- 

Of modern references, \ 
Cahen, La Syrie du Nord a I'epoque des Croisades el la 
principaute franque d'Anlioche, Paris 1940; J. 
Sauvaget, Alep: essai sur le developpement d'une grande 
ville syrienne des origines au milieu du XIX' siecle, Paris 
1941; Ulla S. Linder Welin, Sayfal-dawlah's reign in 
Syria and Diydrbekr in the light of the numismatic evidence, 
in Sueca Reperlis, i, Lund 1961; H. Busse, Chalifund 
Grosskonig, Wiesbaden 1968; and esp. Canard, 
Miscellanea orientalia, Variorum Reprints, London 
1973, L'expansion arabo-islamique et ses repercussions, 
ibid, London 1974, and art. hamdanids. These 
may be supplemented by J. Bacharach, The career of 
Muhammad b. Tughj al-Ikhshid..., in Speculum (197 5); 
P. de Smoor, Kings and Bedouins in the Palace of Aleppo 
as reflected in Madam's works, Manchester 1985; H. 
Kennedy, The Prophet and the age of the Caliphates, 
London and New York 1986; Th. Bianquis, Le 
pouvoir politique a Alep au VI XV siecle, in REMMM 
(1992), no. 62, 49-59; M. Gil, A history of Palestine, 
634-1099, Cambridge 1992, esp. 320. 

(Th. Bianquis) 
SAYF al-DIN, the honorific title of two 
members of the Shansabanid or Ghurid [q. v.] dynasty 
which ruled in Afghanistan and adjoining lands 
during the 6th-early 7th/12th-early 13th centuries: 

1 . Sayf al-DIn Surf b. c Izz al-DIn Husayn, succeed- 
ed his father as ruler in Ghur 540-3/1145-8, killed in 
battle with the Ghaznavid Bahram Shah [see 

GHURIDS. 1. at 11, 1100]. 

2. Sayf al-DIn Muhammad b. c Ala> al-DIn Hu- 
sayn, ruler in the Ghurid capital of FTruzkuh [q.v.] 
556-8/1161-3 [see ibid, at II, 1101]. 

SAYF al-DIN BAKHARZI, Abu 'l-Ma c ali Sa c Id 
b. Mutahhar b. Sa'id b. C A1I (586 : 659/l 190-1261), 
known honorifically as Shavkh-i c Alam and, more 
familiarly, as Kh w adja-yi Fathabadi, in reference to 
the Bukharan suburb of Fathabad where he estab- 
lished a khanakah, a leading disciple of Nadjm al- 
Din Kubra (d. 618/1221), eponym of the Kubrawl 
order [q.v.]. 

After elementary education in his birthplace of 
Bakharz, a town in the Kuhistan region of Khurasan. 
Sayf al-DIn studied jurisprudence and the recitation 
and exegesis of the Kurgan in Harat and Nishapur 
before proceeding to Kh w arazm, the seat of Nadjm 
al-DIn Kubra. According to Nizam al-DIn Awliya 3 
(cited in Amir Hasan Sidjzl, FawdHd al-fu^dd, Buland- 
Shahr 1275/1855, 268-9), Bakharz! was initially 
hostile to Kubra, and even to Sufism as such, but this 
is not confirmed by any early source. Even before go- 
ing to Kh w arazm, Bakharzi had received at least one 
Sufi cloak of initiation, from a certain Shaykh Tadj al- 
DIn Mahmud in Harat, and there can be no doubt 
that he went to Kh w arazm expressly to join Kubra's 
circle. Thanks to an exemplary degree of devotion, he 

advanced swiftly in the esteem of the master. In 
celebration of his bedding a newly-acquired con- 
cubine, Kubra once instructed his disciples to 
dispense with their usual austerities and select some 
pleasurable activity instead; the indulgence chosen by 
Bakharzi was to stand all night outside Kubra's 
chamber with a pitcher of water ready for his master's 
post-connubial ablutions. By way of reward, Kubra 
prophesied that one day rulers would run respectfully 
alongside Bakharzi' s horse, a prediction that is said to 
have come true. Not long after, Kubra interrupted 
Bakharzl's second quadragesimal retreat to tell him 
that his training was complete, and he dispatched him 
to Bukhara in order to propagate the Kubrawl path 
( c _Abd al-Rahman DjamI, Nafahat al-uns , ed. Mahmud 
c AbidI, Tehran 1370 Sh.l\99\, 433-4). 

Bakharzi was to spend the rest of his life, amoun- 
ting to some forty years, in Bukhara; his only 
absences were annual visits to Shaykh Nur al-DIn 
Baslr in Samarkand, timed to coincide with the ripen- 
ing of the highly esteemed khalTli grapes of the region 
(Muhammad b. c Abd al-Djalll Samarkandl, Kandiyya, 
ed. Iradj Afshar, Tehran 1367 Sh./1988, 98-9). He 
soon gained considerable prestige and influence, and 
his eminence was acknowledged by Sufis of other 
lineages, such as Kh w adja Gharlb and Hasan 
Bulgharl of the Kh»adjagan (Fakhr al-DIn C A1I Sail, 
Rashahat c ayn al-hayal, ed. C A1I Asghar Mu'inlyan, 
Tehran 2536 imperial/1977, i, 54-5). When 
Sorkaktani, mother of Mongke, the Mongol Great 
Khan, donated 100 silver balish for the construction of 
a madrasa in Bukhara, despite her own allegiance to 
Christianity, it was to Bakharzi that she entrusted its 
supervision, as well as the administration of the en- 
dowments she settled on it ( c Ata Malik Djuwaynl, 
Tankh-i Djahdn-gusha, ed. Kazwlnl, iii, 8-9). 
Sorkaktani further provided for the establishment of 
Bakharzl's khanakah at Fathabad; the statement of 
Ma'sum C A1I Shah (d. 1344/1926) that the khanakah 
was built by Tlmur in 788/1386 must be taken to refer 
to a restoration or expansion {TaraHk al-hakaHk, ed. 
Muhammad Dja c far Mahdjub, Tehran n.d., ii, 242- 
3). Bakharzi evidently regarded himself as obliged, by 
virtue of his prominence, to influence the Mongol 
rulers in favour of Islam; this is suggested by the ver- 
sified letter he addressed to Kutb al-DIn Habash 
c AmId, vizier to Caghatay Khan, one line of 
which reads, "You are entrusted, in this government, 
with promoting the truth (nusrat-i hakk); should you 
fail to do so, what will be your excuse on the Day of 
Gathering?" (cited in V.V. Bartol'd, Turkestan v 
epokhu mongol'skogo nashestviya, in Socineniya, i, 
Moscow 1963, 541). Bakharzi was visited in Bukhara 
by Berke, the future ruler of the Golden Horde, and 
either converted him to Islam or strengthened him in 
the affirmation of the faith (J. Richard, La conversion 
de Berke et les debuts de Vislamisation de la Horde d'Or, in 
REI, xxxv [1967], 173-84). For all his prestige, 
Bakharzl's zeal for Islam sometimes led him into con- 
flict with the Mongols; thus he was once abducted 
while praying to the Mongol camp outside Bukhara 
and detained for a while" (Abu '1-Mafakhir Yahya 
Bakharzi, Awrad al-ahbdb wa-fusHs al-ddab, ed. Iradj Af- 
shar, Tehran 1358 &./1979, 270). Similarly, one of 
his disciples, Burhan al-Din Bukharl. sent to 
Kubilay's capital of Khan-Ballk to propagate Islam, 
found himself expelled to Macln, i.e. the realm of the 
southern Sung, where he soon perished (Kh w an- 
damlr, Habib al-siyar, ed. DjalSl al-DIn Hunia'I, 
Tehran 1333 Sh./195i, ii, 64). 

Bakharzl's influence extended beyond Bukhara to 
Kirman. Kutlugh Turkan KhatGn, the Kutlugh- 


Khanid ruler of Kirman (r. 658-81/1258-82) re- 
quested him to send her one of his sons, and he ac- 
cordingly dispatched Burhan al-DTn Ahmad (d. 
696/1297), his middle son, bearing with him the gift 
of a tooth allegedly from the Prophet. The khdnakdh 
she established for him in Kirman became a centre for 

Kubrawl order in southern andTouth-eastern Persia. 
The prestige Bakharzi thus acquired in Kirman is 
reflected in the panegyric kasida addressed to him by 
Kh w adju KirmanI (Diwan, ed. Ahmad SuhaylT 
Kh w ansarl, Tehran 1336 $71957, 598-600). Bakhar- 
zi sent his youngest son, Mazhar al-DIn Mutahhar, to 
Konya, not, however, to spread the Kubrawl path, 
but to pay homage to Djalal al-DIn RumI, samples of 
whose verse had reached Bakharzi from a follower in 
Shlraz (Ahmad AflakI, Mandkib al- c drifin, ed. Tahsin 
Yaztci, Ankara 1959, i, 143-5). 

Bakharzi's khdnakdh at Fathabad flourished for at 
least a century after his death and burial there in 
659/1261. His eldest son, Djalal al-DIn Muham- 
mad — the only son to have remained in Bukhara — 
cannot have played a significant role in this regard, 
for he was killed a mere two years after the death of 

s father 

in the . 

ie of fa 


Mongols. It was rather Abu '1-Mafakhir Yahya (d. 
736/1335-6), the son of Burhan al-DIn Ahmad, who, 
coming to Bukhara from Kirman in 712/1312-3, suc- 
ceeded in consolidating the affairs of the khdnakdh 
(Ahmad b. Mahmud Mu c In al-Fukara', Tdrikh-t 
Mulld-zdda dar dhikr-i mazdrdt-i Bukhdrd, ed. Ahmad 
Gulcln-i Ma c anl, Tehran 1339 Sh./1960, 43). Still- 
extant documents from 726/1326 and 734/1333 record 
an augmentation of the endowments settled on the 
khdnakdh; of particular interest is the earmarkini 

e pure 

and n 

mportance, i 

ained in Kirman, being honorifically 
e of them perpetuated the Bakharzi 

belonging to the wakf (O.D. Cekhovic, Bukharskie 
dokumentl XIV veka, Tashkent 1965). Ibn Battuta 
visited the khdnakdh during Abu '1-Mafakhir's tenure; 
he remarks on the lavish hospitality which he received 
and on the singing of poems in Persia and Turkish by 
the assembled dervishes {Rihla, iii, 27-8, tr. Gibb, iii, 
554). The esteem in which rulers continued to hold 
Bakharzi is illustrated by the burial in his proximity 
of the Caghatayid Buyan (or Bayan) Kull Khan in 
760/1359 (E. Knobloch, Turkestan, Munich 1973, 
212). Indeed, gifts by rulers continued to enrich the 
wakf as. late as the 12th/18th century, but by the time 
the Persian dervish Ma'sum C A1I Shah visited 
Fathabad in 1316/1898-9, the shrine was in an ad- 
vanced state of dilapidation, many of its tiles having 
been stolen for sale in the markets of Bukhara {TardHk 
al-hakaHk, ii, 242). 

Long before this architectural decay set in, the 
Bakharzi line of the Kubrawiyya had vanished from 
Bukhara, supplanted — like all other Central Asian 
branches of the order — by the Nakshbandiyya. The 
tomb of Bakharzi came to function simply as a site for 
popular pilgrimage (as happened also with the shrine 
of Nadjm al-DIn Kubra in Kuhna Urgandj). 
Genealogical descendants of Bakharzi can be traced in 
Bukhara as late as the 13th/19th century, and in 
1255/1839 one of them, c Abd al-Kayyum Khodja, 
wrote a biography of his great forebear (Mandkib-i Say/ 
al-Din Bakharzi, mss. Institute of Oriental Studies of 
the Academy of Sciences of the Uzbek Republic, 
6965, 10,802); none of them, however, were practi- 
tioners of the Kubrawl path. A similar development 
took place in Kirman. Burhan al-DIn Ahmad's 
khdnakdh fell into disuse after Abu '1-Mafakhir's 
departure for Bukhara, and although other members 

The only extended prolongation of the Bakharzi 
branch of the Kubrawiyya took place in India. From 
Badr al-DIn Samarkandl (d. 716/1316) went forth a 
line that crystallised after two generations as the Fir- 
dawsiyya of Bihar, a sub-order, of which the most 
eminent representative was Sharaf al-DIn Manerl (d. 
782/138 [see makhdum al-mulk sharaf al-din]) 
(S.A.A. Rizvi, A history of Sufism in India, Delhi 1978, 
i, 226-40; B.B. Lawrence, Notes from a distant flute: Sufi 
literature in pre-Mughal India, Tehran 1978, 72-8). 

Bakharzi's surviving literary corpus is quite 
meagre. It consists of WakdY al-khalwa, an Arabic ac- 
count of visions experienced while in Kh w arazm. on- 
ly a fragment of which is extant (Brockelmann, S I, 
810); a Persian treatise on love {Risdla-yi 'Ishk, ed. 
Sa c Id Naflsl, in Madjalla-yi Danishkada-yi Adabiyat-i 
Ddmshgdh-i Tihrdn, viii/4 (1340 Sh./I96l), 11-24); and 
a number of quatrains, including some of uncertain 
attribution (S. Khuda Bakhsh, Saijuddin Bakharzi, in 
ZDMG, lix (1905), 345-54; Sa c Id Naflsl, RubaHyydt-i 
Bakharzi, in MDADT, ii/4 (1334 Sh./1_955), 3-15; 
Dhablhullah Safa, Tarikh-i Adabiydt dar Iran, Tehran 
1339 S./1951, ii, 856-8). Bakharzi also kept ajournal 
(ruz-ndma) in which he recorded such matters as the 
gifts that were brought to him and the prayers he 
made on behalf of the donors; the adversaries in 
Bukhara for whose redemption and guidance he 
prayed; and the menstrual cycles of his wives and con- 
cubines. The journal is lost but portions of it were in- 
cluded by Abu '1-Mafakhir Yahya in his Awrad al- 
ahbdb wa-fusus al-ddab, an account of the litanies and 
practices of the Bakharzi branch of the Kubrawl 

Bibliography: Iradj Afshar, Say j al-Din Bakharzi, 
in MDADT, ix (1341 S-/1962), 28-74; idem, Saifr 
al-din Bakharzi, in W.B. Henning and E. Yarshater 
(eds.), A locust's leg: studies in honour of S.H. Ta- 
qizadeh, London 1962, 21-7; D. DeWeese, The 
eclipse of the Kubraviyah in Central Asia, in Iranian 
Studies, xxi/1-2 (1988), 47-50; F. Meier, Die FawdHh 
al-Gamdl wa-Fawdtih al-Galdl des Nagm ad-Din Kubra, 
Wiesbaden 1957, 42-3; J. Paul, Scheiche und Herrscher 
im Khanat Cagatay, in Isi, lxvii (1990), 282-3. 

_ (Han 



sodist and minor poet at the TImurid court in 
Harat during the second half of the 9th/15th century. 
He is remembered for his text-book of Persian pro- 
sody '■Arud-i Sayfi, which he completed in 896/1491; 
this has been published several times in India, notably 

tary in H. Blochmann's The prosody of the Persians ac- 
cording to Saifijami, and other writers, Calcutta 1872, a 
work which played an important role in making Per- 
sian poetical theory accessible to European students. 
But now that older and more detailed works on the 
same subject are available (especially the early 
7th/13th-century Mu'-dfam of Shams-i Kays) Sayfl's 
largely derivative work is only of limited interest. 

"Sayfi" was the pen-name of several other Persian 
poets, the earliest being C A1I b. Ahmad al-Sayfl al- 
Naysaburl, whose poems are quoted by c AwfI, 
Shams-i Kays and Djadjarml and of whom Dawlat- 
shah says that he flourished under the Kh w arazm- 
shah c Ala 3 al-DIn Tekish (568-96/1172-1200). 

Bibliography: Sayfi c ArudI is mentioned in Mir 
C A1I Shir Nawa'I's Maajalis al-nafaHs, Babur's 
autobiography, Dawlatshah and later authorities. 


See also Rieu, Persian cat., 525-6; Khayyampur, 
Farhang-i sukhanwaran, 284; Storey, iii/1, 185-7 and 
(for Sayfi NaysaburT) Storey-de Blois, v/2, 516-17. 

(F.C. de Blois) 
SAYFI HARAWI (Sayf b. Muhammad b. 
Ya'kub), poet and historian of Harat in the 
Mongol period. Bom in ca. 681/1282, he gained ac- 
cess to the court of Fakhr al-DIn Muhammad (d. 
706/1307), the third malik of the Kart [q.v.] dynasty of 
Harat, in whose honour he claims to have written 80 
kasidas and 150 kifas. In 706/1306, when Harat was 
besieged by the Mongol army of the Il-Khan 
Oldjeytii, led by Danishmand Bahadur, Sayfi com- 

al-Din Muhammad Sam, who conducted a vigorous 
defence on behalf of the absent malik. On the city's 
fall, Sayfi narrowly escaped execution by Danish- 
mand Bahadur's son Bodjey, whose father had been 
treacherously killed by Sam; and for a time his for- 
tunes underwent an eclipse. It was some years before 
he gained the favour of the new malik, Ghiyath al-DTn 
Muhammad (d. 729/1328-9), to whom he had 
presented a work on ethics, the Madjmu c a-yi Ghiyathi. 
and who commissioned him to write a chronicle of 
Harat. Sayfi's Ta\ikh-i Harat (or Ta\tkh-i muluk-i 
Kart), his only surviving work, covers the period from 
the city's capture by Chinggis Khan in 618/1221 
down to 722/1322, in all probability the date of com- 
position. It was originally intended to comprise two 
daftars totalling 400 discourses (dhikrs); but only the 
first daftar, of 138 dhikrs, has come down to us. 
Although Sayfi's chronology for the first half-century 
or so must be treated with caution, the Ta^rikh, as the 
earliest extant history of Harat, is extremely valuable: 
Sayfi cites lost sources, notably the Kart-nama of 
Rabrn-yi BushandjI {ca. 702/1302-3), and utilises 
documents in the Kartid chancery. Both Hafiz-i 
Abru, in his historical works and in his geography, 
and Isfizari (d. 903/1498), the later chronicler of 
Harat, made extensive use of Sayfi's TaMkh. 

Bibliography: Storey, i, 354-5; Storey-Bregel', 
ii, 1042-4; Abdul Muktadir, Notes on a unique history 
of Herat, in JASB, N.S., xii (1916), 165-84; I. P. 
Petrushevskiy, Trud Seyfi, kak islocnik po istorii 
vostocnogo Khorasana, in Trudi Yuzno- Turkmenistanskoy 
arkheologiceskoy kompleksnoy ekspeditsii, v, A shkh abad 
1955, 130-62; edn. of Sayfi, The TaMkh-nama-i 
Harat, by M.Z. as-Siddfqi, Calcutta 1944, repr. 
Tehran J 352/1 973 . (P. Jackson) 

SAYHAN, the ancient Saros, the western twin 
river of the Djayhan [q.v.] in the Qukurova (Cili- 
cian plain) in eastern Turkey; both were identified 
with two of the Kur'anic rivers from Paradise (Kurgan 
XLVII, 15, etc.) by the early Muslim border war- 
riors. Its total length is 560 km/348 miles with the two 
spring rivers, the western Zamanti Irmagi (325 
km/200 miles), originating in the Uzuh Yayla, and 
the eastern Goksu (200 km/125 miles) coming from 
Binboga and the Tahtah mountains. The Zamanti 
flows through the meadows and gorges of the Taurus, 
and below Farasa Koyu (Ariarameia) it unites with 
the Goksu. The latter, in its upper part preserves the 
ancient name "as Sanz Suyu near the town of Sariz. 
The main right-bank tributaries of the Sayhan are the 
Gorgun Suyu and the Cakit Suyu (Hafiz Suyu, flow- 
ing down from Ulukisla through the famous Pozanti 
Deresi) just above Adana [q.v.], the original mouths 
of which are now immersed in the artificial lake of the 
Seyhan Toprak Baraji (begun in 1956). The riverbed 
had meandered through the Qukurova, causing enor- 
mous inundations and often enough mixing with the 
Djayhan. It now flows westwards into a delta, one 

arm of which is the Tarsus Qayi, and into the sea at 
Deli Burnu. Avoiding the bad climate of the flood 
plain, the main harbour of the region, Mersin [q.v.], 
lies 25 km/15 miles to the north. Health and 
agriculture have been greatly improved by the con- 
struction of the dam, though cotton production was 
known in the Cilician plain from the Middle Ages 
and, together with orchards, had already been 
organised on industrial lines under Mislrli Ibrahim 
Pasha {mutasarrif in Adana on behalf of Mehemmed 
C A1I 1833-40) and during the 1860s. The first railway 
connection of Adana with Mersin was built in 1886. 

The only antique monument preserved along the 
river is the old bridge in Adana with 14 of its original- 
ly 21 arches still in use (length 319 m/1,046 feet), 
erected under Hadrian, with restorations, according 
to inscriptions and literary sources, in the 4th century, 
under Justinian (Procopius, Deaedif., 5 § 5); under al- 
Mahdl in 165/781-2; in 225/840; by Mehemmed IV 
in 1072/1661-2 (Ewliya Celebi, ix, 338); by Ahmed 
III; twice in the 19th century; and in 1949. It is 
wrongly described as having only one arch by Yakut 
(i, 179). It should probably not be identified with the 
Djisr al-Walld [II], constructed between al-Massisa 
(on the Djayhan) and Adana in 125/743 (al- 
Baladhuri, 165), though the latter was situated on the 
western bank. But the region to the west of the 
Djayhan remained a no-man's-land between Arabs 
and Byzantines until the early c Abbasid period, when 
under al-Mansur in 141 or 142/758-9 Adana was 
"founded" (rebuilt) and populated by Khurasanians. 
and in 165/781-2 a fort was begun next to the bridge 
by Harun al-Rashid as governor of the thughur; ac- 
cording to another tradition the city was [rejerected 
only in 194/809-10 (al-Baladhun, 166-7). The early 
history is retailed by Ibn al- c Adim (d. 660/1262, 
Bughya, i, 169-71 , also quoting several hadtths for iden- 
tifying the Sayhan with that river of Paradise having 
water (against three others with honey, wine and 
milk), 381-8, 389-92; cf. al-Mas c udI, Murudj, i, 145 § 
289, ii, 66 § 775; Hamd Allah Mustawfi, Nuzha, ed. 
Le Strange, 209, tr. 201-2, explicitly states that these 
rivers were intended, and not the parallel Sayhun and 
Djayhun of Transoxania). He adds valuable 
geographical and historical information on this region 
(Byzantine occupation, from the end of the 9th cen- 
tury until around 1172, with the intrusion of the 
Lesser Armenian Kingdom of Sis 1080-1375 and the 
Crusaders), and mentions that the river flowed in 
Byzantine territory until Kal c at Samandu (recalled in 
the modern Zamanti, see Honigmann, Ostgrenze, 65- 
6), then in Armenian territory (bildd al-Arman) until 
Adana (i, 379-80). 

From Saldjuk times onwards Turkish tribes occa- 
sionally pushed into the Taurus region, occupying the 
yaylas or summer pastures and valleys; the Ramadan- 
oghullari [q.v.] were later prominent here. Remnants 
of nomadic life have survived down to the recent past, 
and the '■dshiks and popular poets (ozan) of the 19th 
century in this region were famous, including 
Dadaloghlu and Karadjao gh lan. singing of social ten- 
sions and of the natural beauties of the mountains and 

The region and the geography of the river were 
forgotten in Mamluk and Ottoman times, in spite of 
the continuing use of routes via the Cilician pass 
(Giilek Bogazi) and along parts of the Zamanti and 
Sayhan, so that Ewliya Celebi confuses the rivers 
(Adana on the Djayhan, but not consistently, ix, 333, 
336-40, already like the Byzantine sources, see al- 
massIsa). Foreign travellers like Marco Polo in 1270, 
Bertrandon de la Broquiere 1432, Katib Celebi, Paul 


Lucas 1706, Kinneir 1813 and Texier 1835, rarely 
mention more than the route stages and towns. 

Bibliography: For Arabic sources, travellers and 
the main secondary works, see djayhan. Also Ibn 
al- c AdIm, Bughya, ed. S. Zakkar, Damascus 
1408/1988; Ibn Shaddad, al-AHak, Syrie du Nord, ed. 
A.M. Edde, in BEO, xxxii/3 (1980-1), 369, tr. 
Damascus 1984, 91-2; Map: Turkiye 1:800,000, 
Harita Umum Miidurlugu, 1933; L. Rother, Gedanken 
zur Stadtentwicklung in der Cukurova, Wiesbaden 1972 
( = TAVO, Reihe B, 3); Yuri Ansikiopedisi, i, Istan- 
bul 1981, s.v. Adana Hi, 10-57; Murat Yiiksel, 
Cukurova'da Turk-Islam eserleri ve kitabeler, Ankara 
19_94. (C.P. Haase) 

SAYIGH, Fath Allah (1790-?; still alive in 1847), 
son of Anthony, a Christian of the Latin rite who was 
a native of Aleppo and was the clerk, dragoman 
and biographer of Lascaris de Vintemille, 
Napoleon Bonaparte's agent. In Lascaris's company, 
he made a "commercial" tour through different 
regions of Syria and c Irak. This trip concealed what 
was in reality a plan involving high politics, which 
Sayigh was unaware of at the beginning of the travels. 
It was a question of getting to know the desert, its 

its tribes, of placing them under a single chief, of 
organising their rebellion against both the Turkish oc- 
cupying power and also the Wahhabis, and of prepar- 
ing the way for a French military expedition which 
would block the route to India for British commerce. 

Sayigh and Lascaris left Aleppo on 18 February 
1810. They succeeded in reaching Palmyra and in 
placing themselves under the protection of Muhanna 
Fadil, the shaykh of the Hisana. But the policy of 
Muhanna's elder brother, Nasir, was contrary to 
Lascaris's aims. The latter therefore set his sights on 
the amir Duray c b. Sha c lan, the shaykh of the Ruwala 
[q.v.]. It was at that point that he revealed to his com- 
panion the journey's real aim, and Sayigh henceforth 
threw all his energy into realising Lascaris's projects. 
The exploration of the deserts of the Arab East oc- 
cupied the two of them for several years. Sayigh made 
by himself a journey to Dir c iyya [q.v.] in 1813, and 
the Odyssey only came to an end after Napoleon's 

After Lascaris's death at Cairo in 1817, the result 
of being poisoned, the survivor Sayigh went back to 
his mother at Lattakia, where he found a minor job 
with a merchant. He spoke with his friends about his 
numerous adventures and told them of a collection of 
notes which he had put together, at Lascaris's request, 
during his life as a Bedouin. At the time of his travels 
in Syria, Alphonse de Lamartine became interested in 
Sayigh 's memoirs (written in a semi-dialectal Arabic), 
negotiated their purchase, had them translated into 
French and published them in vol. iv of his Voyage en 
Orient (1835 ed.). Sayigh made a trip to Paris in 1847, 
where he was the guest of Lamartine, and thanks to 
the latter, he was nominated consular agent of France 

Studded with errors as it was, Lamartine's version 
of Sayigh's memoirs was condemned to oblivion by 
the Arabists of the 19th century, who accused the 
Syrian traveller of having put together a romance out 
of assorted pieces. It is probably true that one can find 
in Sayigh's account numerous historical and 
geographical errors. But the Journal Asiatique's severe 
judgement, which claims to have reduced the 
historicity of this account to nothingness, is only true 
, in regard to the written parts deformed in translation 
or completely foreign to the original text, as is shown 
on the evidence of the new French version. 

Bibliography: Arabic ms. of Fath Allah Sayigh's 
memoirs, B.N. fonds arabe 2298, ed. as Rihlat Fath 
Allah Sdyigh al-Halabi, ed. Yusuf Shulhud, 
Damascus 1991; Lamartine, Voyage en Orient. Recit 
du sejour de Fatalla Sayeghir chez les Arabes erranls du 
grand desert, new tr. J. Chelhod, Le desert et la gloire, 
Paris 1991; Lettre de Fresnel a Jules Mohl, mJA, 6th 
ser., xvii (1871), 165-83; JA, Rapport annuel, 6th 
ser., xx (1872), 36; Auriant, La vie extraordinaire de 
Theodore Lascaris, Paris 1940; J. Chelhod, Lascaris et 
Sdyigh, agents de Napoleon, chez les Bedouins, in Rev. 
d-histoire diplomatique, 102 1 ' annee (1988), 5-35. 

(J. Chelhod) 
SAYIGH, Tawfik (1923-1971), Christian 
Palestinian Arab poet, born at Khirba (southern 
Syria) as the son of a Presbyterian minister. In 1925, 
his family moved to Palestine, then, in 1948, to 
Beirut. Sayigh was educated at the Arab College in 
Jerusalem and the American University in Beirut 
(B.A. in English literature in 1945), and studied 
literature at Harvard, Oxford and Cambridge. He 
taught Arabic language and literature in Cambridge 
and London. He was the editor in chief of the new 
Beirut cultural magazine Hiwar from 1962 through 
1967. From 1968 until his death from a heart attack 
on 3 January 1971, Sayigh lectured at the University 
of California at Berkeley. 

Sayigh's poetry, which is devoid of rhyme and 
metre, comprises the following volumes: Thalathun 
kasida (1954), al-Kaslda kaf (1960) and Mu'allakat 
Tawfik Sayigh (1963) (all of them republished in al- 
Madpnu^at al-shi<riyya, London 1990). His poetry, 
much of which was originally published in the Beirut 
literary magazines Shi'-r and Hiwar, has been called "a 
supreme example of an early modernity achieved 
because of the poet's particular qualities of vision and 
technique." (Salma Jayyusi, in M.M. Badawi (ed.), 
Modern Arabic literature ( = The Cambridge history of Arabic 
literature), Cambridge 1992, 152). 

His other writings include essays on Arabic and 
English literature and translations of English verse 
(e.g. 50 kasida min al-shiW al-amiriki al-mu^dsir, and 
T.S. Eliot, RubaHyyat arba^ ( = Four Quartets), both 
2 London 1990). 

Bibliography: Issa J. Boullata, The beleaguered 
unicorn: a study of Tawfiq Sayigh, in JAL, iv (1973), 
69-93; idem, The concept of modernity in the poetry of 
Jabra and Sayigh, in Boullata (ed.), Critical perspectives 
on modern Arabic literature 1945-1980, Washington 
D.C. 1980, 263-79, repr. from Edebiyat, ii (1978), 
173-89; Djabra Ibrahim Djabra, Fi dfubb al-usud, in 
Sht c r, xv (1960), 105-17, repr. in idem, al-Humyya 
wa •l-tufan\ Beirut 1982, 28-42; idem, Tawfik 
Sayigh, dughui al-nar wa 'l-djawhar al-sulb, in al-Nar wa 
\-djawhar, Beirut 1975; Mahmud Shurayh, Tawfik 
Sayigh, sirat shd'ir wa-manfd, London 1989. 

(W. Stoetzer) 
SAYMARA, a town of mediaeval Persia, in 
what later became known as Luristan [q.v.], and the 
chef-lieu of the district of Mihradjankadhak. A 
tributary of the Karkha, which flows into the Karun 
river [q.v.], is still today known as the Saymareh. 

The district passed peacefully into the hands of Abu 
Musa al-Ash c ari's Arab troops (al-Baladhuri, Futuh, 
307), and in mediaeval times prospered as a meeting- 
place of Arab, Persian and Lur ethnic elements, apart 
from the devastations of a severe earthquake in 
258/872 (al-Tabari, iii, 1872-3). The valley which 
comprised the district was fertile and could even 
produce dates, but by Hamd Allah Mustawff's time 
(8th/ 14th century), Saymara was falling into ruins 
(which can still be seen today). A notable scholar and 


! the 

i Abu ' 

Saymari, d. 275/888 [q.v. in Suppl., 

Bibliography. Le Strange, Lands of the Eastern 

Caliphate, 202; Schwarz, Iran im Mittelalter, 470-3. 
(C.E. Bosworth) 

al-SAYMARI, Abu 'l- c Abbas Ya c kOb b. Ahmad, 
scholar and writer, probably a Shfr, of whom we 
unfortunately know very little. 

The Shfr authors of the 5th/ 11th century, in par- 
ticular, al-Sharlf al-Radl, compiler of the Nahdj al- 
baldgha [?.».] (d. 405/1015 [q.v.]), and biographers like 
Ibn al-Nadlm (d. 438/1046), al-Nadjashi (d. 
450/1058) and al-TusT (d. 460/1067), do not have a 
single word about al-Sayman or his work. Moreover, 
there were occasions for this. As far as is known, the 
sole piece of information about him which we possess 
is owed to Ibn Abi '1-Hadid, commentator on the 
Nahdi (d. 656/1258 [q.v.]). This leads us to conclude 
that he lived after al-Sharlf al-Radl and al-Tusi but 
before Ibn Abi 'l-Hadld, if indeed he was not the lat- 
ter's contemporary. If this is the case, he must have 
lived in the period of the 5th-6th/l lth- 12th centuries. 

Thanks to Ibn Abi '1-Hadid, we also know of the 
existence of one of his works, the Kaldm '■All wa- 
khutabuhu ("Words and speeches of C A1I"). But this 
work is completely unknown to the older Arab 
biographers. Its title, however, enables one easily to 
deduce its nature, and it was probably a collection of 
'All's discourses and letters. Following other writers 
who came after al-Sharlf al-Radi (cf. Kh. al-Husaynl, 
Masadir Nahdjal-balagha, Baghdad 1966-8, i, 68-92, at 
70), al-Sayman probably wished in his own book to 
fill in some of the gaps left by the compiler of the 
Nahdi. But the extract from his text given by Ibn Abi 
'l-Hadld — a correspondence between C A1I and 
Mu c awiya, apparently at the outset of their quarrel, 
in which the latter invites his opponent to prepare for 
war (Shark Nahdi al-baldgha, Cairo 1965-7, xv, 82)— 
does not appear in the Nahdi; in the present state of 


n bes; 

Bibliography: Given in the 


al-SAYMARI, Abu 'l- c Anbas [see abu *l- c anbas 

AL-SAYMARI, in_Su PP l.]. 

al-SAYRAFI, Muhammad b. Badr, a prominent 
judge in Ikhshidid Egypt. Abu Bakr Muhammad b. 
Badr b. c Abd Allah (or c Abd al- c AzIz) al-Sayrafi was 
born in Egypt in 264/877-8. His father was a client of 
Yahya b. Hakim al-Kinanl and was a prosperous 
money changer (sarrqf) for whom Abu 'Uraar al-Kindi 
composed his Kitdb al-Mawdli. A Hanafi by madhhab 
and a reporter of traditions, al-Sayrafi studied under 
Abu Dja'far al-TahawI as a Hanafi jurist and heard 
traditions from C AH b. c Abd al- c Az!z al-BaghawI in 
Mecca and from other Meccan and Egyptian 
teachers. Ibn c As5kir briefly refers to al-Sayrafi in his 
Ta^rikh madtnat Dimashk, stating that he stayed in 
Damascus for a while and transmitted hadith there, as 
well as in Egypt, on the authority of C A1I b. c Abd al- 
c Aziz al-Baghawi. 

Al-Sayrafi held different judicial posts before suc- 
ceeding to the judgeship of Egypt (Ua' Misr) under 
the Ikhshidids. For a while he was a deputy to 
Muhammad b. al-Hasan b. Abi '1-Shawarib, who had 
been designated as the chief judge of Egypt by the 
c Abbasid caliph al-Radl. According to Ibn Hadjar al- 
c AskalanT, al-Sayrafi himself held the judgeship of 
Egypt on three occasions, from Shawwal 322/October 
934 to Sha c ban 324/July 936, and secondly from Dhu 
'1-Hidjdja 327/October 939 to Safar 329/December 
940. Muhammad b. Tughdj al-Ikhshld appointed al- 
SayrafT for the third and last time to that office in 

Shawwal 329/July 941, a post he held until his death 
in Sha c ban 330/May 942. Abu c Umar al-Kindi 
reports, on the authority of Ibn Ziilak, that Muham- 
mad b. Badr al-Sayrafi wore fine clothes and had a 
sumptuous house in Fustat. 

Bibliography: Kind!, Kitdb al-wuldt wa-kitab al- 
kudat, ed. R. Guest as The governors and judges of 
Egypt, London 1912, 486-7, 488, 489-90; Ibn 
Hadjar al- c Askalani, Rqf al-isr, in The governors and 
judges of Egypt, 533, 540, 542, 543, 546-7, 550, 551, 
552, 557-62, 564, 566, 570, 572, 573; idem, Raf al- 
isr, ed. Hamid c Abd al-MadjId et alii, Cairo 1957- 
61, i, 72, 76, ii, 265-6, 269, 270, 273, 278, 386. 
(F. Daftary) 
SAYRAM (also Sayram, etc.), the name of a town 
in present-day Kazakhstan , some 7-8 miles east of 
Cimkent, on the Arts river, a tributary of the Syr 
Darya. Kashgharl (ed. and tr. Dankoff, ii, 241; 
repeated in the Ta^nkh-i Rashidi, tr. Ross, 171), in the 
earliest reference to it under this name, identifies it as 
the "White City which is called Isbidjab" [see isfidj- 
ab, in Suppl.]. In Kashghari's day (i, 84) its in- 
habitants spoke "both Soghdian and Turkic." It is to 
be distinguished from a city of the same name in 
Eastern Turkistan (located between Kuca and Aksu) 
which, according to local tradition, was founded by 
captives brought thither by the Kalmyks (Barthold, 
rev. of Ta^rikh-i Amniyya, in Socineniya, viii, 213); the 
great Turkish Sufi, Ahmad Yasawi (d. 562/1166-7 
[q. v. ]) was claimed as a native of this town (Barthold, 
Hist, des Turcs d'Asie Centrale, 111). Ch'ang Ch'un, 
who visited "Sai-lan" in the early 1220s, noted that 
its ruler was a Muslim. Subsequent Chinese travellers 
of the Yuan and Ming eras describe it as a well- 
populated Muslim town with extensive agriculture 
(see Bretschneider, Researches, for references). The 
agricultural productivity, based on irrigation canals, 
is remarked on by later Muslim authors as well (see 
Pishculina, Pis'mmnle, 167-72). Rashld al-DIn 
(Djami'-, ed. Romaskevic, i/1, 72-3) places it in a 
listing of Eurasian toponyms between "Talas" and 
"Iblr and Siblr" and cites it (i/1, 101) as one of the 
regions conquered by the legendary Oghuz Khan. He 
also mentions (i/1, 91) it as "Karl Sayram" (Turk. 
"Old Sayram") as an "old and great city", one day's 
journey in length, with 40 gates, inhabited by Muslim 
Turks. It is frequendy noted in Tlmurid (Tlmur gave 
it to his grandson, Ulugh Beg, see Yazdl, ii, 449) and 
Shaybanid sources, with often-shifting overlordship 
according to the fortunes of war (Barthold, Hist, of the 

Bibliography: 1. Sources. E. Bretschneider, 
Mediaeval researches from eastern Asiatic sources, i-ii, St. 
Petersburg, 1888; N. Elias (ed.) and E.D. Ross 
(tr.), A history of the Mughals of Central Asia, being the 
Tarikh-i-Rashidi of Mirza Muhammad Haidar, Dughldt, 
London 1895; Mahmud Kashgharl, The compendium 
of the Turkic dialects (Diwan Luydt at-Turk), ed. and tr. 
R. Dankoff and J. Kelly, Cambridge, Mass. 1982- 
5; Rashld al-DIn, Djami'- al-tawarikh, ed. A.A. 
Alizade, A.A. Romaskevic el alii, 3 vols. Baku- 
Moscow 1957-80; Sharaf al-DIn Yazdl, Zafar-ndma, 
ed. Muhammad c Abbasi, 2 vols., Tehran 

2. Studies. W. Barthold (V.V. Bartol'd), 
Akademik V.V. Bartol'd socineniya (Moscow 1963-77), 
9 vols., in which see Taarikh-i Emenie. Istoriya 
vladeteley Kashgarii, socinenie Mulli Musi, ben Mulla 
Aysa, sayramtsa, izdannaya N.N. Pantusovim ("The 
Ta 3 rikh-i Amniyya. The History of the Rulers of 
Kashgharia, a work of Mulla Musa ben Mulla 
Aysa, the Sayramite, edited by N.N. Pantusov"), 


viii, 213-19; idem, Histoire des Tuns d'Asie Centrale, 
Paris 1945; idem, History of the Semirechye, in Four 
studies on the history of Central Asia, i, Leiden 1962; 
K.A. Pishculina, Pis'mennie voslocnie istocniki o prisir- 
dar'inskikhgorodakh Kazakhstana xiv-xvii vv. ("Eastern 
written sources on the Syr Darya towns of 
Kazakhstan of the XIV-XVII centuries"), in B.A. 
Tulupbaev (ed.), Srednevekovaya gorodskaya kul'tura 
Kazakhstana i Sredney Azii ("The medieval urban 
culture of Kazakhstan and Central Asia"), Alma- 
Ata 1983, 165-77; C.C. Valikhanov, Sobranie socin- 
eniy v pyati tomakh ( = coll. works in 5 volumes), 
Alma-Ata 1984-5. (P.B. Golden) 

SAY'UN, a town in Wadi Hadramawt [q.v.], 
situated about 16 km/10 miles east of Shibam [q.v.] 
and 24 km/15 miles west of Tarim [q.v.] and approx- 
imately 480 km/300 miles north of the port of 
Hadramawt, al-Mukalla [q.v.] (see H. von Wissmann 
and R.B. Serjeant, map of Southern Arabia, Royal 
Geographical Society, 1958). The town was within the 
boundaries of the Fifth Governorate of the People's 
Democratic Republic of Yemen and now in the 
unified Republic of Yemen. Landberg (Etudes sur les 
dialectes de VArabie meridionale, iii, Datinah, Leiden 
1913, 1820) discusses the spelling of the name, Say'un 
and Saywun, etc., concluding that the former is the 
more ancient spelling and that the -un ending is Mahri 
(cf. names like Khudun, Dammun, Kaydun etc.). 

The area is famed for its fertility and, in particular, 
for its date growing. Although it has nothing like the 
renown of Tarim in this regard, some accounts of 
Say'un mention the town as a centre of Islamic 
scholarship (hidjrat Him) (Ibrahim Ahmad al-Makhafi", 
Mu'-djam al-bulddn wa 'l-kaba'il al-Yamaniyya, San'a 3 
1988, 335). Landberg, at the turn of the century 
(Etudes, i, Hadramout, Leiden 1901, 451), writes of the 
temporary decline of Tarim as a centre of scholarship 
and religion with the closure of its school (ribdt) and 
the resultant rise of the discipline in Say'un. Here a 
rich benefactor of the school, al-Habib (the usual 
Hadramf title of a sapid) c Ali al-Habshl Ba c AlawT, 
was its rector, providing for all the needs of poor 

Say'Cin, it should be said, is not of great antiquity. 
The name of the town appears along with the Kathlri 
sultans from about the 9th/15th century. They con- 
quered Hadramawt from the east, from Zafar [q.v.] 
from where they originally hailed. The town con- 
tinued in their hands and was the capital of the 
Kathlri sultanate in 1967 when the British left the 
Aden Protectorates and independence was granted. 
Bibliography : All references are given in the text 
except the historical ones, the main one of which is 
Salim b. Muhammad b. Salim b. Hamld al-Kindl, 
Tdrikh Hadramawt, al-Wdda al-mufida al-ajflmi'a li- 
tawdrikh kadima wa-haditha, ed. c AbdaIlah Muham- 
mad al-Habshi, San'a' 1991, passim. See also 
Government of Bombay, An account of the Arab tribes 
in the vicinity of Aden, Bombay 1909, 123 ff.; R.B. 
Serjeant, The Portuguese off the South Arabian coast, Ox- 
ford 1963, 25 ff. and passim; idem, Omani naval ac- 
tivities off the Southern Arabian coast in the late UM17th 
century from Yemeni chronicles, in Jnal. of Oman Studies , 
vi (1983), 77-89. (G.R. Smith) 

SAYYID, Sa'id (a., pis. asydd, soda, sdddt, abstract 
nouns siydda, suHad, etc.), originally, chief, e.g. of an 
Arabian tribe, and then, in Islamic times, a title of 
honour for descendants of the Prophet Muham- 
mad, being in this respect in many ways coterminous 
with the term sharif. 

Sayyidv/as used in ancient South Arabian, where it 
appears as s'wd "chieftain" (A.F.L. Beeston, etc., 

Sobaic dictionary, Louvain-Beirut 1982, 129), but the 
root seems to be largely absent from North-Western 
Semitic, being only dubiously attested in Elephantine 
Aramaic (J. Hoftijzer and K. Jongeling, Dictionary of 
the North- West Semitic inscriptions, Leiden 1995, ii, 779- 
80). In early Arabic usage, it could be applied to 
animals, with the wild ass called the sayyid of his 
female, and to the leaders of the Jinn (Lane, Lexicon, 

1. In early Arabic tribal usage. 

Though sayyid is the most common and the most 
general term, there are many other designations for 
the tribal chief: c amid, Hmad "pillar, support"; rabi' or 
sayyid al-mirba^ (from the chiefs entitlement to a 
quarter of captured booty); rami's, kdHd (referring to the 
chiefs leadership in war); khatib and zaHm (referring 
to his oratorical powers); and shaykh and kabir 
(designating chiefs who were veterans, with long and 

shaykh has become in recent centuries the standard 
title for the Bedouin tribal chief, since both sayyid and 
sharif (referring to the chief's nobility of lineage) have 
tended to acquire specialised Islamic religious senses 
(see below, 2.). 

The chiefs authority in his tribe was based in the 
first place on his nasab or ancestry, from inheritance 
(ink"') and from a chain of noble (sharif) and free 
(hurr, sarih) predecessors (kabir™ c an kabir'") (see Bichr 
Fares, L'honneur chez les Arabes avant I'Islam, Paris 
1932, 52; and <ird). He had to possess such qualities 
and practice such skills as hilm [q.v.], magnanimity; 
dahd^, shrewdness and diplomatic finesse; and liberali- 
ty and hospitality (a chief might be described as '■azim 
al-kudur "having large cooking pots" or kathir al-rimdd 
"having a large pile of ashes [outside his tent]"). He 
had to act as spokesman and orator for the tribe, so 
that many chiefs were famed also as poets, such as 
Imru 5 al-Kays of Kinda and Malik b. Nuwayra of 
Yarbu c of Tamim [q.vv., and see for other such poet- 
chiefs, sha'ir. 1. a]. Such abilities as being able to 
read and write and being able to swim are recorded as 
being amongst those possessed by the chief of the 
Khazradj in Medina, Sa c d b. Ubada [q. v.]. Because of 
an emphasis on seniority within the leading family, 
chiefs tended to be men of mature age and experience 
rather than youths; they might be bald, but could not 
be beardless. Certain of the chiefs functions might on 
occasion be exercised by others of his kinsmen or by 
tribal notables, and in early modern times, Jaussen 
noted the existence at the side of the chief of a specific 
leader of raids, '■akid al-ghazw, amongst the Jordanian 
tribes (Coutumes des Arabes au pays de Moab, Paris 1948, 

A chief might succeed to the headship by an act of 
bequest (wasiyya) of his predecessor. Imru 3 al-Kays 
succeeded his father Hudjr; Hatim al-Ta'fs [q.v.] son 
succeeded him; and Durayd b. al-Simma [q.v.] suc- 
ceeded his brother. Although chiefs usually came from 
what would later be called a shaykhly house, this 
lineage had to be allied to proven competence if the 
tribe were to flourish and survive in a harsh and 
lawless environment. Hence because of the overriding 
necessity for continuing success, the chiefs authority 
was fragile and could be impaired in times of adversi- 
ty. That of c Amir b. al-Tufayl [q.v.] was challenged 
within his tribe of Dja c far b. Kilab of c Amir b. Sa c sa c a 
after setbacks in battle, but later restored. Unless the 
chief was one of the minority of outstandingly forceful 
figures regarded as uncontested in their authority 
(muld c , ghayr mudafa c ), such as Muhammad's contem- 
porary c Uyayna b. Hisn of Fazara, his authority 
rested to a considerable extent on the tacit consent of 


his peers within the tribe, underpinned by his unflag- 
ging ability to preserve his people's fighting reputa- 
tion and to secure pasture for its herds. Hence chief- 
tainship could often be more a burden than a source 
of exploitation, reflected in the popular saying sayyid 
al-kawm khadimuhum "the chief is the tribe's servant". 
Whether, as Chelhod surmised, the chief originally 
had a sacral role, one effaced by the advent of Islam, 

spokesman one might conjecture an attribution of 
magical powers similar to those attributed to the an- 



Bibliography (in addition i 
tide): G. Jacob, Altbeduinisches Leber, nach den Quellen 
geschilderC, Berlin 1897; H. Lammens, Le berceau de 
I' Islam, I 'Arable occidentale a la veille de I'Hegire, Rome 
1914, 203-331; C.A. Nallino, Sulla costituzione delle 
tribii arabe prima dell'Islamismo, in Racolta di scritti, 
Rome 1939-48, iii, 64-86; E. Tyan, Institutions du 
droit public musulman. i. Le Califat, Paris 1954, 84- 
102; J. Chelhod, Introduction a la sociologie de I'lslam, 
Paris 1958, 53-9; J. Henninger, La societe bedouine 
ancienne, in F. Gabrieli (ed.), L'antica societa beduina, 
Rome 1959, 82-4. (C.E. Bosworth) 

See for this, sharif, especially section 3. 

mad b. Muhammad al-Husaynf al-Ashraf, Persian 
poet who died presumably in 556/1161. He spent the 
greater part of his life in Ghazna as a panegyrist of the 
Ghaznawid Sultan Bahrain Shah (5 1 2-47/1 1 1 8-52), to 
whose campaigns into India he dedicated several 
kasidas. During the latter's reign he made the Hadjdj. 
in all probability prompted by problems with this 
mamduh and intended as a search for a new one. Our 
oldest source on Hasan, the Persian polymath Zahir 
al-DTn Abu '1 Hasan al-Bayhakf [q. v. ], mentions in his 
(Arabic) Lubab al-ansdb (written in 555/1160) that he 
met the poet, who was on his way to Mecca, in 
Nlshapur in 544/1149 (Mudarris-i Radawl, introd. to 
the Diwan, 357). During his journey he associated 
with members of the Saldjuk dynasty: he mourned 
sultan Mas'ud's death in a larajf-band (in 546/1151) 
and dedicated a kasida to the accession to the throne of 
Malik Shah III (in 547/1152). After this journey he 
seems to have resided in Khurasan at the court of San- 
djar. Following Sandjar's capture by the Ghuzz in 
548/1153, he attached himself to the latter's protege 
and nephew, the Karakhanid Mahmud Khan, who 
had become ruler of Khurasan. It is not completely 
certain whether kasidas dedicated to the Kh w arazm 
Shah Atsiz (d. 551/i 156) and to the Saldjuk Sulayman 
Shah on the occasion of his enthronement in 555/1 160 
were only sent to, respectively, Kh w arazm and 
Hamadhan or refer to an actual stay at those places. 
The kasida for Sulayman Shah gives the last date of his 
life, which came to an end before 557/1162, the date 
of the deposition of Mahmud Khan (see below). A 
tomb of the poet can be seen in Azadwar near 

The only work which has survived is a Persian 
diwan of about 5,000 bayts. This Diwan is preceded by 



who a: 

with the author's wish in hi: 
ment, he collected it to dedicate it to the Karakhanid 
Mahmud Khan. Although no ms. of the Diwan older 
than the late 10th/16th century survives, several of 
Hasan Ghaznawl's poems have come down to us in a 
ms. from 636/1238 from another text, namely, 
RawandT's [q.v.] history of the Saldjuks Rabat al-sudur 
(completed in 603/1206-7). In this text he is, after 
Mudjlr al-DIn Baylakam [q.v.], the diwan author most 

cited; i 

mplete poems (kasi 

tarkib-bands), which, in accordan 
the text, are all intended for Saldjuk mamduhs, and 
some odd verses as well. Moreover, Rawandl cites an 
anecdote wherein Hasan advises another poet to learn 
by heart "modern" poets like Anwari [q.v.] and 
himself and leave aside the great classics like Sana'! 
[q.v.] whose talent would inhibit a poet (57-8). 

The Diwan consists of kasidas, strophic poems, 
ghazaliyydl, kita c at and rubaHyyat. The kasidas are mostly 
panegyric, and more than half of them lack a nasib and 
begin directly with the madh. A large part of the re- 
maining kasidas have a nasib devoted to a description 
of spring and taghazzul. Some kasidas, however, stand 
out both by subject matter and quality. Memorable 
are no. 57, a complaint kasida with the title safir al- 
damir; no. 59, which echoes a kasida by Mas c ud-i Sa c d- 
i Salman [q.v.] combining complaint with Jakhr, and 
no. 60, an ascetic kasida. In no. 72, a sawgand-ndma 
("oath-poem"), the poet tries to repudiate an accusa- 
tion. Some kasidas have long passages devoted to was}. 
A case in point is the original no. 46, which has a des- 
cription of fourteen lines of the eyes, which leads to a 
gurizgdh where the gaze is oriented towards the mam- 
duh. Other examples are the description of the night 
in nos. 62 (as well as of the horse) and 67 (with 
separate attention to each of the planets). 

Already in 1958 Braginski paid attention to the 
ghazaliyyat of Hasan. More than a third of them end 
with a panegyrical dedication; in this he follows 
Mas c ud-i Sa c d-i Salman and MuSzzT [q.v.] and 
prefigures his later contemporary Sana 3 ! and 
ultimately Hafiz. 

Bibliography: Diwan, ed. Muhammad Mudar- 
ris-i Radawi, 2nd revised ed., Tehran 1362 Sh. 
(with extensive introduction); Rawandl, Rabat al- 
sudur ed. Muhammad Ikbal, with extra notes by 
Mudjtaba MmuwT, 2 Tehran 1364 Sh.; Muhammad 
c Awfi, Lubab al-albab, ed. Sa c Id NafTsI, Tehran 1335 
Sh., 438-41; Dawlat Shah, Tadhkirat al-shu<ara\ ed. 
Muhammad RamadanI, Tehran 1338 57;., 82-4; 
Lutf- C A1I Beg Adhar, Atashkada, ed. Hasan Sadat 
Nasiri, ii, Tehran 1338 Sh., 535-44; Braginski, 
vozniknovenii gazeli v tadzhikskoy i persidskoy literature, 
in Sov. Vostok, ii (1958) 94-100; Dhablhallah $afa, 
Tarikh-i adabiyyat dar Iran, ii, Tehran 1332 Sh., 586- 
98; J. Rypka, Hist, of Iranian literature, 122, 197; 
idem, in Camb. hist. Iran, v, 562; Munzawi, Fihrist-i 
nuskhahd-yi khatti-yi fdrsi, iii, Tehran 1350 Sh., 2297- 
98; J.T.P. de Bruijn, Of piety and poetry. The interac- 
tion of religion and literature in the life and works of Hakim 
Sana^i of Ghazna, Leiden 1983, passim (see index); 
Khaliki-Mutlak, art. Asraf Gaznavi, in EIr (based 
on Mudarris-i RadawT's introduction). 

(Anna Livia Beelaert) 
al-SAYYID al-HIMYARI, Abu Hashim Isma'Il 
b. Muhammad b. Yaztd b. Rab^a b. Mufarrigh, a 
Shi c i poet and a grandson of the poet Ibn Mufarri gh 
al-Himyan [q. v.]. He was born to IbadT parents about 
105/723, grew up in Basra, and died in Baghdad or 
Wasit between 173/789 and 179/795. At a young age, 
he adopted with great fervour the doctrine of the 
Kaysaniyya [q.v.] Shfa, believing in the imamate and 
occultation of Muhammad b. al-Hanafiyya [q.v.] and 
his return as the Mahdi [q.v.]. Twelver Sh^I 
authorities, both mediaeval and modern, claim that 
he later converted to Imamism (tadfafara), but the 
proof-stories they cite are anachronistic, contradic- 
tory, or legendary, and his ladfafur poems have been 
judged of old as forgeries. Despite his well-known 
Kaysani ShT < i beliefs, though, he maintained good 
relations with the c Abbasid caliphs al-Saffah, al- 


Mansur, a!-MahdI and al-Rashid, whose praises he 
sang and whose gifts he accepted. 

Al-Sayyid is considered a leading and prolific 
muhdath poet of the class of Bashshar b . Burd and Abu 
'l- c Atahiya [q.vv.\. Although several known people 
(including four daughters) transmitted his poems, no 
one could collect them all, due to their large 
number— allegedly over 2,300. Of these, only 221 
poems and fragments have survived; and, of the many 
books written about his akhbtir, only one is extant also. 
Clearly, his poetry was shunned very early, since it 
contained vehement invectives against the Compan- 
ions, the Prophet's wives, and the enemies of the 
Shf-Is, notably^ the Umayyads. Most of his poetry 

and extraordinary exploits. Though this poetry is 
generally lyrical, melodious, easy-flowing, and 
idiosyncratically narrative in style, al-Sayyid's most 
tender are perhaps his Kaysani poems, depicting Ibn 
al-Hanafiyya's state and his long-awaited return to 
earth. These latter poems have been a major source 
for the heresiographers on the Kaysaniyya and have 
often been attributed to the earlier Kaysani poet, 
Kuthayyir c Azza [q.v.]. 

Bibliography: 1. Texts. Diwdn al-Sayyid al- 
Himyari, ed. Shakir Had! Shakar, Beirut n.d.; Ibn 
al-Mu c tazz, Tabakat al-shu c arti\ 6-8, 32-6; Isfahan), 
al-Aghanl, ed. Beirut, vii, 224-68; Kashshi,, 
ed. Karbala 5 , 242-5; Marzubani, Akhbtir al-Sayyid al- 
Himyari, Beirut 1993; al-Shaykh al-Mufid, al-Irshad, 
tr. I.K.A. Howard, Elmhurst 1981, 429-39; 
DhahabI, TaMkh al-isltim, ed. Tadmurl, xi, 157-61; 
idem, Siyar aHtim al-nubala\ viii, 40-2; Kutubi, 
Fawtit al-wafaytil, ed. c Abbas, i, 188-92; Safadi, al- 
Wtiflbi 'l-wafaytit, ix, 196-202; Ibn Hadjar, Listin al- 
miztin, i, 436-8; Kh^ansarl, Rawddl al-gjanndt, 
Tehran 1390, i, 103-11; MadjlisI, Bihdr al-anwir, 
Tehran 1377, x, 232-3, 1393, liii, 131-2. Of the 
heresiographical sources, see Nawbakhti. Firak al- 
SkPa, ed. Nadjaf, 46-7; Kumml, al-Maktiltit wa 7- 
firak, 36-7; [ps.-j al-Ngshp al-Akbar, MasdHl al- 
imdma, 27-9. 

2. Studies. Barbier de Meynard, Le Seid 

Himyarite, mJA, 7eme ser., iv (1874), 159-258; M. 

al-Amin, A<ytin al-sh^a, Beirut 1961, xii, 84-165; 

Brockelmann, I, 83, S I, 133; Sezgin, ii, 458-60; 

W. al-Kadl, al-Kaystiniyya fi '1-taMkh wa 'l-adab, 

Beirut 1974, 322-56; M. Nouiouat, La vie d'al- 

Sayyid al-Himyari, poke chiile du IF I VII' siecle, in REI, 

xlviii (1980), 51-97; H. Halm, Die islamische Gnosis, 

Zurich-Munich 1982, 54, 71; J. van Ess, Theologie 

und Gesellschafl im 2. und 3. Jahrhundert Hidschra, i, 

Berlin-New York 1991, 304-5. (Wadad Kadi) 

SAYYID KUTB, Ibrahim Husayn Shadhill, 

Egyptian writer, prominent member of the 

Muslim Brotherhood and main ideologue of 

modern Muslim Sunnf fundamentalism, born 9 

October 1906 in Musha near Asyut, executed 29 


1966 ir 

Life. In 1920 Sayyid Kutb moved from his native 
village to Cairo for his secondary education. From 
1929 till 1933 he studied at Ddr al- c Ulum. He worked 
as a teacher for approximately six years, became a 
functionary in the Ministry of Education (Wizarat al- 
Ma'-tirif), and was sent on an educational mission to 
the United States where he spent two years. He 
returned to Egypt in August 1950. 

Sayyid Kutb joined the Muslim Brotherhood [see 
AL-IKHWAN al-muslim0n] probably only after his 
return from the United States. The exact date on 
which he joined is still disputed. It is usually assumed 
that he became a Brotherhood member in 1951. 

In October 1952 he left the Ministry of Education 

after nineteen years of service. When the July 1952 
revolution took place, he was in close contact with the 
Free Officers. He served, so we are told by several 
sources, as the cultural adviser to the leaders of the 
revolution. Nevertheless Sayyid Kutb parted with the 
new leaders because of ideological differences, for he 
believed that Islam should serve as the basis for the 
new Egyptian regime. 

He became the editor of the newspaper of the 
Muslim Brotherhood just before he was arrested for 
the first time, together with the leadership of the 
Muslim Brotherhood, in early 1954, and remained 
behind bars for three months. 

His second arrest took place on 26 October 1954 in 
the wake of the Manshiyya incident, supposedly an 
attempt to assassinate the then President, Djamal 
<Abd al-Nasir [q.v. in Suppl.], in Alexandria's Man- 
shiyya Square. In 1955 he was sentenced to fifteen 

Thanks to mediation by the c Iraki president c Abd 
al-Salam c Arif, Sayyid Kutb was released from prison 
in 1964. In November 1964 he published his Ma c tilim 
fi 'l-tarik "Landmarks". In this book he accuses pres- 
ent Muslim societies of being not Islamic but djahili. 
This word originally simply means "pre-Islamic", 
but in this new context it has no limitations to a period 
but means first of all "pagan", "barbaric", "anti- 
Islamic", "vicious" and "wicked". The accusation 
of being djahili implies apostasy from Islam, which is 
punishable by death. 

On 9 August 1965, Sayyid Kutb was arrested 
again. He was accused of attempting to assassinate 
Djamal c Abd al-Nasir, of treason, and of planning a 
coup d'etat. His trial was presided over by the 
notorious military judge Fu'ad al-Didjwi who, in the 
words of Gilles Kepel, "offered the accused all the 
guarantees of fairness characteristic of a military court 
in a dictatorial state trying defendants broken by tor- 
ture." Sayyid Kutb was sentenced to death on 21 
August 1966 and executed one week later. 

Autobiographical writings. Two of Sayyid Kutb's 
novels are generally believed to be largely 
autobiographical: Tifl min al-karya "A child from the 
village", 1946, and Ashwdk "Thorns", 1947, a mov- 
ing work which explains why its hero never married. 

At the end of the year 1965, in prison, Sayyid Kutb 
wrote a detailed statement on his activities within the 
Muslim Brotherhood. This statement was published 
under the title Li-mtidhti a'-damuni? "Why did they exe- 
cute me?". Moreover, several letters, autobio- 
graphical articles, and court or police proceedings 
have been preserved. 

Literary works and literary criticism. The first book 
Sayyid Kutb published was his Mahammat al-shaHr fi 7- 
haytit wa-shi c r al-djjl al-hadir, "The task of the poet in 
life and the poetry of the present generation", 1933. 

Sayyid Kutb's first and last volume of poetry ap- 
peared in 1935. It is entitled al-ShtitP al-madjhill, "The 
unknown shore". The articles of literary criticism 
which he published in the periodicals al-Ristila, al- 
Muktataf al-Kitab and al-Ktitib al- Misri between 1942 
and 1946 have been collected in his Kutub wa- 
shakhsiyydt, 1946. In the same year his novel al-Madina 
al-mashura "The enchanted city", appeared. 

Writings on Islam. Ideas. Sayyid Kutb's most impor- 
tant work is without doubt his commentary on the 
Kur'an, Fizilal al-Kur^tin, literally "In the shadows of 
the KurW, in 30 volumes. When Sayyid Kutb was 
arrested in 1954, 16 volumes had appeared. The re- 
maining volumes were written from prison, with 
Shaykh Muhammad al-Ghazall acting as govern- 
ment-appointed censor. 

In 1960, a revised edition of this commentary 


started to appear. Volume xiii of the revised edition 
(up to sura XIV) appeared in 1964. Arrest and execu- 
tion prevented the revised edition from being com- 
pleted. It is this revised edition which is often 
characterised as tafsir haraki "activist exegesis". 

In the general introduction to this commentary, the 
layman Sayyid Kutb carefully explains that he "had 
heard God speaking to [him] in this Kur'an", even 
though his training had been literary, non- 
theological, and even though he had never been a 
recognised Azhar-trained "man of religion". 

His al-'Adala al-iajtimaHyya fi 'l-Isldm, "Social 
justice in Islam", April 1949, revised ed. 1964, was 
originally written as an answer to the leftist ideas with 
which Egypt was permeated in the late nineteen- 
forties. It contains much that would come back in 
more rigidly-argued forms later on. It appears to be 
the first work in Arabic that employed the phrase 
"social justice" instead of "socialism". 

The dfahiliyya theory developed in Ma c dlimfi 'l-tarik, 
1964, became the ideological nucleus of modern Sun- 
nl fundamentalism. It has the grim consequence of 
takfir, the act of identifying someone as a kafir, 
unbeliever, or, even, when born a Muslim, as an 
apostate who deserves the death penalty. 

Sayyid Kutb's djahiliyya theory has roots in tradi- 
tional Islam and in traditional Islamic law. It appeals 
to the traditional dislike which the inhabitants of the 
Middle East feel for their rulers. It reflects the feeling, 
common in the Middle East, of being overwhelmed by 
a modernity that penetrated the Muslim world from 
the West. 

There is, however, more to fundamentalism than a 
mere rejection of modernity. The non-fundamentalist 
Sufis, as well as other groups, equally reject moderni- 
ty, but they do not accept the grave political conse- 
quences which the fundamentalists attach to this re- 

Amongst the collections of articles that appeared as 
separate books under the name of Sayyid Kutb, 
Ma'-rakaiund ma c a 'l-Yahud, "Our struggle with the 
Jews", deserves individual mention. The essay which 
gave this volume its title was probably written in 
1950-1. In its present form, it contains footnotes refer- 
ring to the spurious document known as The Protocols 
of the elders of Zion that were added in 1970 by a Saudi 
editor, one Zayn al-DIn al-Rakkabl. There is an 
English translation of the essay by Ronald L. Nettler, 
including an elaborate introduction and commentary, 
see Bi hi. below. 

A complete list of Sayyid Kutb's writings can be 
found in SaJah <Abd al-Fattah al-Khalidl, 1991, 517- 
80. This valuable list supplies ample information on 
the origins, first date of appearance, reprints etc. of 
Sayyid Kutb's articles and books. 

Bibliography: Salah c Abd al-Fattah al-Khalidl, 
Sayyid Kutb: min al-mildd ild 'l-istishhdd, Damascus, 
Beirut-Jeddah 1991, 608 pp. (discusses the primary 
and secondary sources); Sylvia G. Haim, Sayyid 
Qutb, in Asian and African Studies, xvi (1982), 147-56; 
Yvonne Y. Haddad, Sayyid Qutb: ideologue of Islamic 
revival, in J. Esposito (ed.). Voices of the Islamic 
Revolution, New York 1983. 

Translations into English of works by Sayyid 
Kutb: Syed Qutb Shaheed, Milestones, tr. S. Badel 
Hasan, Karachi 1981; tr. anon. Cedar Rapids, 
Iowa n.d.; Islam and universal peace, Indianapolis 
1977; R. L. Nettler, Past trials and present tribulations: 
a Muslim fundamentalist's view of the Jews, Oxford 
1987, repr. 1989. 

Texts by Sayyid Kutb tr. into French: O. Cam 
and G. Michaud, in Les Freres Musulmans, Paris 
1983, 83-107. 

All books on modern Islam discuss Sayyid Kutb, 
but see especially: G. Kepel, The Prophet and 
Pharaoh, London 1985, 26-67 (Gilles Kepel, Le Pro- 
phete et Pharaon: les mouvements islamistes dans I'Egypte 
contemporaine, Paris 1984, 30-67); E. Sivan, Radical 
Islam, New Haven 1985, 21-28; J.O. Voll, Fun- 
damentalism in the Sunni Arab world, in M.E. Marty 
and R. Scott Appleby (eds.), Fundamentalisms ob- 
served, i, Chicago 1991, 368-72. 

(J.j.G. Jansen) 
SAYYIDS, a dynasty of Indo-Muslim kings in 
Dihli which followed the Tughluks and preceded the 
Lodis [?.ot. ] and ruled over Dihli for about 37 years 
(817-55/1414-51). Four rulers, Khidr Khan (817- 
24/1414-21), Mubarak Shah (824-37/1421-34), 
Muhammad b. Farid (837-47/1434-43) and <Ala> al- 
DIn c Alam ghah (847-55/1443-51), belonged to this 
dynasty. Their claim of Sayyid descent seems to have 
been shrewdly fabricated in order to buttress their 
position in the absence of any racial or oligarchic sup- 
port. The contemporary author of TaMkh-i Mubarak- 
Shahi, 182, gives two reasons for this claim: firstly, 
because Sayyid Djalal al-DIn Bukharl, an eminent 
SuhrawardI saint of Multan, had once referred to 
Khidr's father, Malik Sulayman, as a Sayyid; and 
secondly, because Khidr possessed the moral qualities 
of a Sayyid. Both are flimsy arguments. Significantly 
enough, the author of TaMkh-i-Muhammadi is silent on 

Khidr Khan, the founder of the dynasty, was the 
son of Malik Sulayman, an adopted son of Malik 
Naslr al-Mulk Mardan Dawlat, an influential amir of 
Flruz Shah Tughluk. Under the later Tughluks, 
Khidr became the mukfa' of Multan, but was removed 
in 798/1395-6 when he quarrelled with Sarang Khan, 
brother of MallG Ikbal Khan [q.v.]. Fortune smiled 
upon him when Timur occupied Dihli and conferred 
its government on him. On the eve of his departure, 
Timur entrusted to him the government of Multan 
and Dipalpur also. This gave Khidr an initial leverage 
over other Tughluk maliks who were struggling for 
supremacy in the state, but he had to struggle hard to 
reach the throne. After a number of battles and skir- 
mishes at Adjodhan, Sirhind, Rohtak, Mewat and 
Dihli and in the Do'ab, Khidr entered Sin on 17 
Rabi c I 817/6June 1414 and ascended the throne. His 
seven years of rule were spent mostly in quelling 
rebellions and dealing with recalcitrant groups of the 
nobility. He made determined efforts, though without 
any lasting success, to reunite under the Sultanate the 
whole tract of the country from Multan in the west to 
Kannawdj in the east and from the foot of the 
Himalayas in the north to the borders of Malwa. He 
was, however, reluctant to assume sovereignty and 
was content with the title of Rdyat-i aHa, pretending to 
rule on behalf of Timur. Abu '1-Fadl remarks: 
"Khidr Khan in gratitude (to Timur) did not assume 
the regal title but styled his court 'the Sublime Stan- 
dards', and adorned the khutba with the name of that 
illustrious monarch and afterwards with that of Mirza 
Shah Rukh. but it concluded with a prayer for 
himself" (ii, 312). The text of thekhutba is given in the 
Maajma* al-insha' and Ta^rlkh-i Muhammadi. Khidr 
Khan died on 17 Djumada I 824/20 May 1421 and, 
according to Firishta, people mourned him by wear- 
ing black garments. 

Mubarak Shah, who succeeded his father, dealt 
with the rebellions of Djasrath Khokar and Tughan 
Ra'is, garrisoned Lahore and undertook campaigns 
against Katehr and Kampil. Developments in Etawa, 
Multan, Bayana, Gwaliyar and Mewat disturbed him 
throughout his reign. He also had to deal with Sultan 
Ibrahim Shark! of Djawnpur [q.v.], undertake cam- 

paigns against the Khokars and face the revolt of 
the sons of Sayyid Sallm (Shawwal 833/June-July 
1430). In Djumada II 834/February-March 1431, 
he had to deal with the incursion of Shaykh c Alf of 

On 9 Radjab 837/19 February 1434, Sultan 
Mubarak Shah was assassinated when he was prepar- 
ing to go for his Friday prayers. During the 17 years 

of his reign, he had ti 

operations in Katehr, Mewat and the north-eastern 
region, and had to deal with rebellious maliks in and 
around the capital. Corrupt and treacherous officers 
added to his problems and pervaded the entire ad- 
ministrative machinery. His administrative short- 
comings apart, he was (as Firishta says) a "cultured 

Immediately on his father's death, Muhammad 
Shah (son of Farid b. Khidr) ascended the throne 
"with the assent of the amirs, maliks, imams, Sayyids, 
grandees, people, 'ulamd* and the kadis" (T.-i 
Mubdrak-Shdhi, 236). The next day he called the high 
amirs and slaves of Mubarak, who enjoyed the 
privilege of the mahi maratib (fish banner), on the 
pretext of giving bay c at (allegiance), and had some of 
them killed and put others in confinement; he thus 
brought many of the regicides to book. At a time when 
anarchical tendencies were getting out of control, 
some <ulama> and amirs invited Mahmud I KhaldjI 
[q.v.] from Malwa. In utter distress, Muhammad 
Shah sent for Bahlul Lodi and his troops from 
Samana. The stern battle of the first day convinced 
Mahmud that the conquest of DihlT would not be an 
easy task. He therefore readily accepted Muhammad 
Shah's proposal for peace and turned back. When he 
was on his return journey, Bahlul Lodi plundered 
some of his equipage. Pleased at this, Muhammad 
Shah addressed Bahlul officially as his son. 

In 845/1441 Muhammad Shah handed over 
Dipalpur and Lahore to Bahlul and commissioned 
him to chastise Djasrath Khokar. But Bahlul turned 
hostile and marched against Dihli. This undermined 
the position of Muhammad Shah, whose nobles 
"even within twenty karohs of Dihli " turned against 
him. On Muhammad Shah's death in 847/1443, his 
son '■Ala 5 al-DIn c Alam Shah was raised to the throne. 
According to Abu '1 Fadl, he "possessed no share of 
rectitude and abandoned himself to licentious 
gratifications" (ii, 313). Reports of SharkI attacks and 
the disloyal attitude of the nobles created a difficult 
. In 1448 he abandoned Dihli and 

iettled i 

r. Bahlul Lodi took possession of Dihli but al- 
lowed him to rule over Bada'un till his death 
(883/1478). Subsequently, Sultan Husayn SharkI in- 
tegrated Bada'un into his kingdom of Djawnpur. 
Families which traced their origin to him were known 
in Bada'un as Khidr-Khani Sayyids (Fadl Akram Sid- 
dlkl, Athar-i Bada>Un, Bada'un 1915, 70). 

Emerging as the principality of Multan, the Sayyid 
dynasty ended as the principality of Bada'un. Looked 
at in the context of the Dihli Sultanate, the Sayyid 
dynasty forms a watershed in the history of mediaeval 
India, indicating a stage in the dismemberment of 
centralised monarchy. Its rulers were devoid of any 
ideal of establishing an empire; their political vision 
was confined to a radius of some 200 miles round 
Dihli. Their writ worked only from "Dihli to Palam" 
(az Dihli ta Palam), as the saying went. They under- 
took innumerable punitive campaigns but these were 
mainly directed against their own maliks. There was 
neither administrative coordination nor uniformity in 
the areas under their control. Even in the agrarian 

and collection" (Moreland, 67). Group assessment 
seems to have gained ground at the expense of Shar- 
ing or Measurement. In reality, the Sayyid kings of 
Dihli— with their nebulous title of rayat-i aHa— were 
nothing more than glorified ikta c -dars and, as Tripathi 
has remarked, "were never seriously considered as 
Sovereign rulers". Their political outlook, their 
theory of kingship, and even their ethnic origin, was 
born of the exigencies of the situation. The 
numismatic evidence on which E. Thomas and 
Nelson Wright based their conclusions about inscrip- 
tions on the Sayyid currency is not borne out by infor- 
mation recorded by Bihamid Khanl and other Persian 
chroniclers. According to Bihamid Khanl, the orders 
of Shah Rukh were enforced here for about 40 years. 
Robes of honour and standards were received from 
Harat, while the sikka (currency) was issued and the 
khutba was recited in the name of Tfmur and his son. 
It appears from the TPrikh-i Muhammadi and the 
Matla' al-sa'dayn that the Sayyid rulers regularly paid 
tribute to them. Mubarak Shah's claim to be Nd'ib-i 
Amir al-Mu'minin should be interpreted in the context 
of Shah Rukh's "ambition of being recognised as 
Khalifa and overlord of other Muslim princes" 
(Arnold, The Caliphate, 12). 

Despite all the political turmoil and unstable condi- 
tions that prevailed in their shrunken kingdom, the 
Sayyid rulers displayed a keen interest in the founding 
of new cities. Khidr Khan founded Khidrabad and 
Mubarak Shah Mubarakabad on the banks of the 
Djumna (Sir Sayyid Ahmad Khan, Alhar al-sanddid, 
ed. Khallk Andjum, Delhi 1990, i, 102). No ar- 
chaeological remains of these cities have survived. 
c Ala 3 al-DIn is said to have populated A c lapur in the 
Bada'un district (RadI al-DIn Bismil, Kamz al-la\ikh, 
Bada'un 1907, 232). He laid out a pleasure garden 
also (Manzoor Badaoni, Athar awliyd-i shahr-i Badd'un, 
Agra 1338/1920, 25). The architecture of the Sayyid 
period is, however, expressive of dissolution. Only the 
tombs of the period have survived; it has been rightly 
designated as the Makbara architecture. According to 
Percy Brown, 26, "in the course of time the country 
around the capital was converted into a vast 
necropolis". More than fifty tombs of considerable 
size and importance belonged to nobles. No large con- 
gregational mosques were built during this period. 
Bibliography: 1. Sources. Yahya Sirhindl, 
TaMkh-i Mubdrak-Shdhi, Bibl. Ind., Calcutta 1931; 
Mir Kh w and, Rawdat al-safd, Lucknow 1270- 
4/1853-7; Hindu Shah, cailed Firishta, TaMkh-i 
Firishta, Nawal Kishore 1281/1864-5; Sharaf al-DIn 
Yazdi, Zafar-ndma, Bibl. Ind., Calcutta 1885-8, ii; 
Haydar b. Abu '1 C A1I Ewoghlu, Maajma c al-insha 71 
B.L. Or. 3482, fols. 38a-39a; Nizam al-DIn, 
Tabakat-i Akbari, i, Calcutta 1927; Bihamid Khanl, 
TaMkh-i Muhammadi, B.L. Or. 137, fols. 311b- 
312a; Abu '1 Fadl, A'in-i Akbari, tr. H. Blochmann, 
repr. Delhi 1989, ii, 312; c Abd al-Kadir Bada'unl, 
Muntakhab al-tawarikh, Bibl. Ind., Calcutta 1864-9; 
c Abd al-Razzak Samarkandl, Matla'- al-sa'dayn, ed. 
M. Shafi c , Lahore 1941-9; Kh»urshah b. Kubad 
al-Husaynl, Ta\ikh-i Kutbi, ed. S.M.H. Zaidi, 
Delhi 1965, 186-7. 

2. Studies. E. Thomas, The chronicles of the 
Pathan Kings oj Delhi, London 1871; N. Wright, The 
coinage and metrology of the Sultans of Delhi, Delhi 1936; 
P. Brown, Indian architecture (Islamic Period), repr. 
Taraporewala 1968; W.H. Moreland, The agrarian 
system of Moslem India, Cambridge 1929; Cambridge 
history of India, iii, ed. Sir Wolseley Haig, Cam- 
bridge 1928; K.S. Lai, Twilight of the Sultanate, rev. 


ed., Delhi 1980; Aziz Ahmad, Studies in the Islamic 
culture in the Indian environment, Oxford 1964; R.P. 
Tripathi, Some aspects oj Muslim administration, 
Allahabad 1936; T.W. Arnold, The caliphate, Ox- 
ford 1924; A. Halim, Administrative system oj the 
Sultans oj Khifr Khan! (the Sayyid) dynasty, in JAS 
Pakistan, iv (1959), 49-72; idem, Local administration 
oj the Sayyid dynasty, in JAS Pakistan, v (1960), 35-62; 
idem, Relation oj the Sayyid kings with the rulers oj pro- 
vincial dynasties, in Procs. oj the Indian History Congress, 
1945, 118-26; idem, Growth and development oj Hindi 
literature during Sayyid-Lodhi period, \r\JAS, Pakistan, ii 
(1957), 69-89; idem, Growth of Urdu language and 
literature during Sayyid— Lodhi period, in JAS, Pakistan, 
iii (1958), 43-66; A comprehensive history of India, v, 
ed. Habib and Nizami, Delhi 1970, ch. VI on the 
Sayyids by K.A. Nizami, 630-40. 

(K.A. Nizami) 
SAZ, Ottoman form saz, the Turkish lute. Saz, 
with various meanings in Persian, including "musical 
instrument" in general, may be applied to stringed 
instruments (the cogur in Adharbaydjan. the viol, 
sorud, in Balucistan), to wind instruments (the zurna in 
Persia) or to the musical band itself; in effect, in 
Balucistan (where the local language often reflects the 
ancient stages of Persian), this word also means the 
tuning of instruments. 

If in Turkey saz still denotes, in a rural environ- 
ment, any musical instrument, the term is generally 
applied to the lute with a long neck and a semi- 
pyriform belly, with three groups of strings compris- 
ing two or three strings, and this has become over the 
course of the 20th century a real symbol of national 

1. History and structure oj the instrument. 
There are representations of lutes with long necks 
on Akkadian cylinder seals of the third millennium 
B.C. The same type of instrument was spread, in the 
second millennium B.C., all over western Asia and 
Egypt. Later, there are Byzantine pictures of long- 
necked lutes (pandoura), attesting the presence of such 
instruments in Anatolia before the Turkish conquest. 
The question of the origin of these lutes can be dis- 
cussed at length, but it is clear that, although such 
stringed instruments may have existed in Western 
Asia before the Turks' arrival, it nevertheless seems 
that the Oghuz brought into Asia Minor their own 
lute, the ancestor of the present saz, with the name of 
kopuz. In descriptions of the 14th-15th centuries A.D. 
(in Yunus Emre and the Dede Korkut epic), it seems 
to have been a lute with three strings, with a long neck 
and a sound-board of hide. The present form of the 
saz could thus be the result of a synthesis between the 
autochthonous, Anatolian instrument and the 
Turkish kopuz. 

In our time, the saz has several names, correspond- 
ing to different sizes respectively; thus, in increasing 
order, cura (between 50 and 70 cm), cura-baglama or 
tambura (ca. 95 cm), baglama (ca. 110 cm), bozuk (115- 
20 cm), divan sazi (130 cm) and meydan sazi (140 cm or 
more). The length of the neck compared with the 
sound-board is such tha'. one can play an octave and 
a fourth. One should mention a final variant, original- 
ly from eastern Turkey and Adharbaydjan, 
characterised by a shorter neck (an octave + a tone) 
and with a total length of about 100 cm; this is called 
cogur, or kisa sapli (baglama). It is not impossible that 
the most archaic saz may be the little cura with three 
strings, called in the western parts uctelli baglama or 
kopuz, and by the nomads of the Taurus, bulgan. 

Each fret is a nylon thread (in former times a piece 
of gut or a copper thread), wrapped three times round 
the neck and tied at the back; this may be the origin 

of another name currently used for the saz, sc. baglama 
(from baglamak "to attach"). 

The strings, of metal since the 18th century, are in 
origin two (ikitelli, corresponding to the Persian dotar) 
or three (uctelli, setar). On the biggest saz, the cords are 
doubled or tripled in unison, in order to increase the 
dynamic or to create sympathetic resonances (dorttelli, 
altitelli, etc.). At the present time there is added a 
deep-toned threaded string, tuned to an octave. The 
strings at the top and at the base are of a similar trim 
( . 1 8 or . 20 mm) , and that or those of the middle is/are 
lower (.30 mm). In our context, we shall speak about 
the lower, middle or top string in the singular, even 
though a group of strings may be involved. 

2. Evolution and playing techniques. 

It is important to distinguish instruments of rural 
origin from those of musical instrument workshops of 
the towns. The latter, meant for playing in all styles, 
have 24 frets and can sometimes claim the same 
perfection as the classical lutes. A saz made by a peas- 
ant or nomad, with 10 to 18 frets, will be adapted to 
a particular regional or tribal style. 

In the countryside, the saz is still played with the 
fingers, with trimmed finger nails; with one finger, or 
with rasgueados; but a plectrum may also be used, now 
made of a piece of plastic with an equivalent 

The different sizes of instruments, the placing of the 
frets and the numerous strings of the three groups 
display the main traits of the Anatolian musical 
system. At the outset, the longer the instrument, the 
more the left hand is forced to make long movements, 
played on the first string, with the other strings acting 
as a drone. The saz is in this case essentially a monodic 
instrument, subordinated to the voice which it accom- 
panies. If it is of small or medium size, it will be 
played on the three strings; the slenderness of the neck 
allows in effect the thumb to block the upper string, 
whilst the other fingers press indifferently the lower or 
middle string. If the hand grips round the neck, the 
third finger may even press the upper string at the 
same time as the thumb. From this there results the 
possibility of varying the drones and of producing 
chords of three sounds, so that the art of the rural saz 
is often characterised by a tendency to polyphony, 
which sustains an invariably monodic melody. 

As many as 14 different tunings (dixzen) of the saz 
are enumerated. The ones most used today are: 

(from top to bottom) 

G-D-A (kara dixzen or bozuk dixzen) 

E-D-A (asik duzeni or baglama duzeni) 

The first allows all the modes of rural Anatolian 
music to be played, and leaves the choice between 
several base notes (A, G, B or D). For this reason, it 
is considered as the most suitable to represent the 
logical and systematic coherence put forward by the 
teachers in the conservatories of traditional music in 
contemporary Turkey. The second, the tuning of the 
minstrels (asik [see c ashik]), is more limited since it 
only allows one base note (E), but is nevertheless very 
esteemed, and considered as more "hot" (yanik). 
Since the base note is situated on the middle string, 
the polyphonic tendency mentioned above is 
displayed there necessarily, in the shape of chords of 
parallel fifths very characteristic of minstrels' art. 

Other tunings are meant for a more limited number 
of modes, and are practiced either in a special region 
(e.g. miskel duzeni, F#-D-A, in the region of Ankara) 
or by a group (e.g. abdal duzeni, G-A-A, characteristic 
of the Abdal Turkmens). 

3. Ideology and society. 

In present-day Turkey, after the coming of modern 
ways of transmitting music — radio, tape recorders — 


and taking into account the more frequent movements 
of population and migration to the big towns, several 
practices of the saz exist together. 

Despite the change brought about by modern life, 
for marriage celebrations, saz musicians of the village, 
knowing the traditional and local repertoire, are still 
resorted to. They often play in small groups e.g. saz, 
viol and a wind instrument. They are gradually being 
replaced by youths playing the urban repertoire on an 

The saz, then, continues to accompany the singing 
of the asiks or ozans, [q. v. ] . These poet-musicians come 
either from the mainly Sunni regions of Kars and Er- 
zurum, or from the Alevi-Bektasi communities of the 
Sivas and Erzincan regions. In this latter case, inter- 
nal emigration, to Istanbul or Ankara, or external, to 
Germany, has perpetuated the tradition of the asik, 
which owes its popularity in the urban milieu to the 
political echoes of the texts of the songs, as expressions 
of a minority often persecuted in Ottoman history, 
and to a social ideal and art of living. Furthermore, 
the saz is the essential instrument for the ceremony of 
the Alevi sama'- [q.v.]. For all these reasons, it has 

porary Turkey and in the diaspora, in the form of the 
cogiir tuned according to the baglama or asik diizeni. 

In addition to these practices, of a local or religious 
nature, since the first years of the Republic there has 
come into being a national music, whose diffusion has 
been assured by the Turkish radio and the state con- 
servatoires. The repertoire is collected from all the 

framework, the saz embodies the national identity in 
more than one way: at the outset by its suitability for 
representing all the regional styles; then by the fix- 
edness of its system fretting and the proportions which 
dertermine its making; and finally, by the inherent 
system and logic in learning the practice of the saz and 
often underlined by its teachers; these last give pride 
of place to the kara duzen, in fifths, assimilated to the 
"rational" order. From this point of view, one can 
say that the two tunings mentioned above are seen to- 
day as entailing two distinct ideological contents: the 
baglama diizeni is associated with the voice of the asik, 
to what is secret and to the heterodoxy of Bektashism 
[see bektashiyya], or even with social confrontation, 
in the eyes of official circles, who set up against it the 
"rational" kara duzen, the expression of the logical 
laws of the national music. 

Bibliography : H.G. Farmer, Studies in oriental 
musical instruments, London 1931; Sir Laurence 
Picken, Folk instruments of Turkey, Oxford 1975; The 
new Grove dictionary of musical instruments, London 
1984, iii, 319-20; M. Stockes, The Arabesk debate, 
Oxford 1992; see also tunbur, c ud. (J. Cler) 
SEBASTIYYA [see sabastiyya]. 
SEBKHA [see sabkha]. 

SEBUKTIGIN (Tkish. sevuk te'gin "beloved 
prince"), Abu Mansur, Turkish slave com- 
mander of the Samanids [q. v. ] and founder of the 
Ghaznawid dynasty [q.v.] in eastern Afghanistan. 
What little is known of his early life stems mainly 
from his alleged Pand-nama or testament of advice to 
his successor (preserved in a later Persian historian; 
see shabankara'I) and from Djuzdjani's quotations 
from a lost part of the Mudjalladat of Abu '1-Fadl 
Bayhaki [q.v.] which dealt with Sebiiktigin's gover- 
norship. He came from the Barskhan district of the 
Semirecye [see yeti su], was enslaved and taken to 
™ i, thus beginning a career in the Samanid 

slave of Alptigin [q.v.]. After Alptigin 


°at Gha'zn' 

, [q.v. 

n periphery of the Samanid 
of his fellow- 

Turkish commanders succeeded him as governor 
there, culminating in Sebuktigin's assumption of 
power in 366/977, beginning a twenty years' reign, 
nominally as governor for the Samanids, in fact as an 
independent ruler. From his base at Ghazna, he 
began the policy of expansion which his son Mahmud 
was to continue, adding Bust and Kusdar [q. vv. ] to his 
kingdom, launching expeditions towards north- 
western India against the Hindushahi [q.v. ] king 
Djaypal and annexing the Kabul river valley down to 
Peshawar. He gave military aid to the last Samanid 
amirs against their rebellious generals (384-5/994-5), 
but just before his death in 387/997 himself intervened 
in Transoxania and established his influence in the 
capital Bukhara. He was succeeded in Ghazna, after 
a brief succession struggle, by his eldest son Mahmud 

After his death, Sebuktigin acquired a reputation 
amongst the Ghaznawids as the just ruler, amir-i c ddil, 
but little of his real personality in lact emerges from 
the sources. 

Bibliography: For the sources, see Bosworth, 
Early sources for the history of the first four Ghaznavid 
sultans (977-1041), in IQ, viii (1963), and for the 
Pand-nama, M. Nazim, The Pand-Ndmah of Subuk- 
tigin, in JRAS (1933), 605-28. Ol studies, see Bar- 
thold, Turkestan down to the Mongol invasion, 261-5; 
M. Nazim, The life and times of Sultan Mahmud of 
Ghazna, Cambridge 1931, 28-33; Bosworth, The 
Ghaznavids, 36-44. (C.E. Bosworth) 

SEFARETNAME [see safTr]. 
SEGBAN [see Suppl.]. 
SEGESTAN [see sistan]. 
SEGOVIA [see shakubiyya]. 
SEGU or Segou, a town of the present Mali, 
situated on the banks of the Niger and the historic 
capital of the Bambara kingdom of Segu from the 18th 
century onwards, and then of the empire of al-Hadjdj 
c Umar. 

There exist four villages with the name Segu, all on 
the river's right bank, at more than 200 km/124 miles 
to the north-east of Bamako. These are, from 
upstream to downstream, over a distance of a dozen 
kilometres, Segu-Koro ("old Segu"), Segu-Bugu (the 
village of agricultural cultivation), Segu-Kura ("new 
Segu") and Segu-Sikoro, transformed into a royal 
residence by Ngolo Jarra (ca. 1750-87). 

It was around 1710 that Biton Mamari Kulibali 
asserted his power, as war leader and founder of a new 
system of authority, over the Bambara peoples, who 
were sedentaries, animists and users of the Mandingo 
language, of this part of the Niger valley. Having 
become the leader of a group of young men, all in the 
same age band (ton), he built up his power through the 
extensive acquisition of slaves from all origins, who 
became "slaves of the ton" (tonjon). In this way, a cen- 
tralised, military state, based on Segu, developed, 
which spread out in all directions. Over two centuries, 
two dynasties succeeded each other at the head of this 
empire, the Kulibali and the Jarra. 

The basic social institution of the Bambara empire 
of Segu was the tonjon or group of subject persons, 
who made up the army and a good part of the 
Through loyal i 

:ould a 

mand. It was these tonjons who, in the middle of the 
18th century, exterminated the Kulibali family and 
raised to power one of their own number, Ngolo 
Jarra, founder of the new dynasty. 

The bases of Segu' s prosperity rested on the lasting 


alliance between an industrious (cotton, indigo, food 
production) Bambara peasantry, the military regime 
of the ton jon, who accumulated captives and booty, 
and the Marka, a specialist Muslim merchant group, 

operations. The Bambara themselves were strongly 
attached to animist cults (the dynasty protected a 
whole network of temples and priests), but tolerated 
the practice of Islam in the quarters of the town- 
enclaves of the Marka. 

At the approach of al-Hadjdj c Umar, in the midst 
of the 19th century, the ruler of the time, Bina c Ali 
Djarra, tried in vain to ally with the Dina of Masina, 
the neighbouring Islamic Fulani power. Al-Hadjdj 
'Ulnar entered Segu on 26-7 Sha'ban 1277/9-10 
March 1861, and proceeded immediately to the 
gathering-together of all the idols and their destruc- 
tion. Designated as khalifa by his father, Ahmadu 
(Ahmad al-Kablr), 'Umar's son, took over the succes- 
sion on his father's death in 1864. It was at this time 
(1864-6) that the French traveller Mage lived in Segu, 
of which he has left a lively and well-informed descrip- 
tion, estimating the population of Segu-Sikoro at that 
time at 10,000 persons. Ahmad al-Kablr reigned in 
this town for almost thirty years, at grips with the op- 
position of some of his brothers and with the con- 
quered population. In this same period, the French 
advanced progressively from Senegal. On 6 April 
1890 Colonel Archinard entered Segu, whilst 
Ahmadu fled eastwards. In the course of the ensuing 
months, Archinard sent to the Bibliotheque Nationale 
in Paris— which has preserved them ever since— the 
rich library of al-Hadjdj c Umar and of Ahmadu. 

Bibliography: E. Mage, Voyage dans le Soudan oc- 
cidental (Senegambie-Niger), Paris 1868; C. Monteil, 
Les Bambara de Segou et du Kaarta, Paris 1924, 2 1976; 
L. Tauxier, Histoire des Bambara, Paris 1942; A. 
Bime, Segou, ville capitate, Angouleme 1952; S. 

de Segou (XVIIF et XIX Steele), diss. Paris 1965; L. 
Kesteloot et alii, Da Monzon de Segou, epopee bambara, 
Paris 1972; G. Dumestre, La geste de Segou - lexles des 
griots bambara, diss. Paris 1974; J. Bazin, Guerre et 
servitude a Segou, in C. Meillassoux, L'esclavage en 
Afrique precoloniale, Paris 1975; S.M. Mahibou and 
J.-L. Triaud, Voila ce qui est arrive. Baydn ma waqa'-a 
d'al-Hajj Wmar al-Futi, Paris 1983; N. Ghali et alii, 
Inventaire de la bibliotheque '■umarienne de Segou, Paris 
1985; D. Robinson, La guerre sainte d'al-Hajj Umar, 
Le Soudan occidental au milieu du XIX Steele, Paris 
1988; F. Simonis, Des Francois en Afrique. Les Euro- 
pe'ens de la region de Segou, diss. Paris 1993. 

O.-L. TRtAUD) 

SEHI BEY, Ottoman poet and biographer of 
poets of the 10th/16th century, b. 874/1470-1 and 
died 955/1548-9. 

1. Life. 

His original name is unknown, and the historian 
c AIT dubbed him " c Abd Allah" and considered him as 
stemming from the Dewshirme [q. v.]. He was certain- 
ly from Edirne and a close associate of the poet 
NedjatI Bey (d. 914/1508-9 [?.».]), also from that 
town. Already when Bayezld's son Mahmud (d. 
913/1507) was appointed governor of Manisa, Sehl 
was the companion of NedjatT and several other poets 
in the Prince's entourage (see Latlff, Tedhkire, Istan- 
bul 1314, 196, 329; c Ashik Celebi, ed. Meredith 
Owens, London 1971, fols. 130b-131a; M art. s.v. 
(F. Akiin); Mehmed Qavusoglu, Necati Bey Divani'nin 
tahlili, Istanbul 1971, 11; Fa'ik Reshad, Tedhkire-yi 
Sehi, cd. Mehmed Shukn, Istanbul 1325, 76, 41; 
Giinay Kut, Heft bihift, the Tezkire by Sehl Beg. An 
analysis of the first biographical work on Ottoman poets, with 

a critical edition based on Ms. Suleymaniye Library, Ayasofya 
0. 3544, Sources of Oriental Languages and Liter- 
atures 5, Cambridge Mass. 1978, pp. 423, at p. 1). 

Through a misunderstanding of 'All's Kunh al- 
akhbdr, Ewliya Celebi wrongly states that Sehl was 
Nedjati Bey's son-in-law, and that he married Nedjati's 
daughter in order to lay claims to his poems after 
Nedjati's death (Seyahat-ndme, Istanbul 1314, i, 343, 
347), a mistake later repeated by F. Babinger in his 
EI' art.; it was corrected by Reshad, in op. cit., 316, 
see also Akiin, op. cit. , and Kut, op. cit. , 2. Also, Bab- 
inger, following Leunclavius and Sidjill-i < Othmani, iii, 
115, also wrongly states, according to Akiin and Kut, 
that Sehi was the secretary (katib) of Bayezld's 
youngest sons Mehmed, and accompanied this last to 
Kaffa [q. v. ] , where he became for a time sandjak beg; 
most sources, however, connect Sehl with Prince 
Mahmud and Manisa (see above). Sehl's close con- 
nection with NedjatT continued till the latter's death, 
and on Sehl's own tomb is a chronogram couplet of 
his mourning Nedjati's death (see Akiin and Kut). 

Later, Sehl served as a secretary to the prince 
Suleyman in Manisa and also with the latter at Edirne 
when Suleyman was commander there for Rumelia. 
After Siileyman's accession as sultan, Sehl recalled in 
his poems former favours no longer vouchsafed, but 
he seems in fact, from other poems, to have remained 
in favour with the royal family, and held office in the 
Morea [see mora]. Finally, he served as administrator 
(mutewelli) of various Hmdrets in Edirne and Ergene, 
dying at the former place; it was whilst acting as ad- 
ministrator of a wakf there he wrote his Tedhkire (see 

2. Works. 

Encouraged by the Kadi c Asker Muhiyy al-DIn 
Celebi Fenari [see fenarT-zade], Sehl put together his 
poems into a dtwdn, the unique ms. of which is in 
Paris, B.N. suppl. turc 360. This contains kasidas, 
ghazah and other lesser genres of poetry, but Sehl does 
not seem to have been considered in his own time as 
a noteworthy poet. This may explain why he is hardly 
quoted in the anthologies of the 10th/ 16th and 
llth/1 7th centuries and why only one ms. of his diwan 
survives. On the basis of chronograms in some poems, 
Akiin concluded that Sehl completed the collection be- 
tween 942/1535 and 944/1537. The dedicatees of his 
poems include the sultans from Bayezid II to 
Suleyman and a large number of Grand Viziers, and 
there are three poems (naH) dedicated to c Ali b. Abi 
Talib and two to Sufi dedes, indicating a possible con- 
nection of Sehi with a dervish order. 

Sehl's real fame sprang from his Tedhkire, finished 
in 945/1538 and called Hesht bihisht "Eight- 
paradises", the first literary biographical work in Ot- 
toman literature (for detailed discussion, see Kut, op. 
cit. , 17-18). Sehl states in his preface that he followed 
the examples of the collections of biographies by 
DjamI, Dawlat Shah and C A1I Shir Newa 3 !, and 
following their example, divided his own work into 
eight labakas. These include (1) the life of Sultan 
Suleyman; (2) the other sultans and the princes who 
wrote poetry; (3) high officials like viziers, nishdndjis, 
etc.; (4) poets from the '■ulema''; (5) the poets who had 
died by the time Sehi was writing (for which he seems 
to have gathered his own research; he has valuable 
details about early poets such as Ahmedi [see 
ahmadT], Ahmed-i Da c T and Sheykhl [q.v.]); (6) poets 
whom Sehl knew personally in his youth, some still 
alive; (7) his contemporaries and the newcomers (in- 
cluding two poetesses); (8) young and talented poets 
who had just started on their careers. An epilogue 
eulogises Sultan Suleyman. 

Following his Persian and Caghatay Turkish 


predecessors, Sehi gives concise information about his 
poets, with full details of names and education, but 
rarely giving dates of birth or death; he then ends with 
selections from his subject's poetry. This procedure 
was adopted as a model by future biographers. His 
work clearly fulfilled a need as pioneer of the genre, 
for shortly afterwards came the similar works of 
Latifi, c Ash!k Celebi and several others. 

There exist 18 mss. of the Tedhkire scattered 
through the library collections of Turkey and Europe 
(see Kut, op. cit., 16-37). Ms. Ayasofya 3544 is the 
basis for Kut's edition; it was probably presented to 
Sultan Siileyman and was subsequently owned by 
Prince Mehmed. An earlier print was issued by 
Mehmed ShukrI, Istanbul 1325/1907, with the title 
Athdr-i eslafdan tedhkire-yi Sehi (but his printed version 
contains only 218 poets, see Kut, 12-14), to which is 
appended a study on Sehi by Fa'ik Reshad; this print 
was based on ms. Millet, Ali Emiri, Tarih 768, copied 
by C A1I Emiri himself. Finally, the Tedhkire was 
translated into German by Necati Lugal and O. Reser 
as Sehi Bey's Tezkere. Turkische Dichterbiographien aus dem 
16. Jahrh., Tubingen 1942. Kut's critical edition is 
based on six mss.; see her Hest bihist and also her Heft 
Bihist 'in yeni bir niishasi ve bir diizeltme, mjnal. of Turkish 
Studies, vii (1984), 243-301. In recent years, Dr. 
Miijgan Cumbur and a group of scholars have been 
working on a serial edition of all Ottoman bio- 
graphical works, starting with Sehl's. 

Bibliography: Given in the article. For the older 

bib]., see F. Babinger's EP art. (G.A. Tekin) 

SELAMLIK (t.), the Ottoman Turkish term for 
the outer, more public rooms of a traditionally- 
arranged house, used e.g. for the reception of 
guests and non-family members; it thus contrasted 
with the inner rooms which constituted the haram or 
harem for the womenfolk. The term seldmlik da>iresi is 
also found. A further use of the word seldmlik is in the 
expression seldmlik dldyi to denote the sultan's 
ceremonial procession from the palace to the mosque 
for Friday worship, a practice kept up by the Ot- 
tomans up to and including Mehemmed V Reshad 
[q.v. ] in the second decade of the 20th century. 
Bibliography: Pakalin, iii, 153-5. (Ed.) 

SELANIK, the Ottoman Turkish name for 
classical and early Byzantine Thessalonike, modern 
Greek Thessaloniki, conventional form Salonica; 
the largest city of Macedonia, on the gulf of the 
same name, to the east of the Vardar river mouth. 
The city has always possessed a large and secure port, 
and was located on the Via Egnatia connecting 
Durazzo (Durres) with Byzantium. In the 5th/llth 
century, it is first named Salonikion, from which all 
variant names derive: Salunik or Salunik in Arabic, 
Solun in Bulgarian, Selanik in Turkish and Salonica 
in English. In the 6th/10th century, the town was an 
important centre of Mediterranean trade, with ties to 
the Islamic world; but apart from al-ldrlsl, the 
mediaeval Arab geographers do not mention it. An at- 
tack on the city by a naval force based upon Tripoli 
in Syria (289/902) supposedly netted the attackers a 
total of 22,000 captives. In 581/1185 Salonica was 
taken by the Normans; the textile artisans, for whom 
the town was famous, were transferred to the royal 
workshops in Sicily. In the early 7th/13th century, 
Salonica was ruled as an independent kingdom by the 
Marquis of Montferrat; but after the reconquest of 
Constantinople by the Byzantines, the Palaeologi 
recovered Salonica as well. Serbian conquests in the 
area resulted in the isolation of the Byzantine exclave 
of Thessaloniki by the middle of the 8th/14th century, 
which was then linked to Constantinople only by sea. 

Ottoman activity in the area began under Murad I, 
with nomads from the western Anatolian principality 
of Sarukhan [q.v.] settling in the area. Ottoman forces 
once conquered the city, but returned it to the 
Emperor Manuel. Sultan Yildirim Bayezid recon- 
quered it in 796/1394, but after his defeat and capture 
in the battle of Ankara (804/1402), his son Suleyman 
returned it to the Byzantines (805-6/1403). Many 
details of this sequence remain unclear. However, 
after the siege of Constantinople by Murad II in 
826/1423, the governor of Thessaloniki, Andronikos 
Palaeologos, sold the city, which then supposedly held 
about 40,000 inhabitants, to the Venetians. While the 
sultan recognised this transfer in the capitulations 
granted to the Venetians in 830-1/1428, in 833/1430 
he conquered the city nonetheless. In the meantime, 
many inhabitants had abandoned the city because of 
the prevailing insecurity. Johannes Anagnostes, a 
Byzantine chronicler, has left a detailed account of 
these events. He claims that 7,000 persons, including 
himself, were taken prisoner. Yet in some cases, the 
sultan himself paid the ransoms of the captives and 
promised that those who had fled the city would have 
their properties restored in case they returned. Two or 
three years later, Turkish settlers were brought into 
Selanik from Yeiiidje-i Vardar, and the church of the 
Acheiropoietos and the monastery of the Prodromos 

A tax register (tahrir) was also prepared at this time, 
but has not survived. We do, however, possess a tax 
register dating from 883/1478 and a fragment from 
the reign of Bayezid II (r. 886-918/1481-1512) 
(Basbakanhk Arsivi Osmanh Arsivi, Istanbul Tapu 
Tahrir 7, and Bibliotheque Nationale Cyrillos and 
Methodios, Sofia, Oriental section, SN 16/35, publ. 
in Bistra Cvetkova (ed.), Fontes turcia histonae 
bulgancae, xvi, Sofia, 1972). There exist two further 

967-8/1560 and another from 1022/1613 (Tapu 
Tahrir 403 and 723). The earliest tahrir enumerates 
862 Muslim and 1,275 Christian householders. From 
their regular distribution among the pre-existing town 
quarters, it can be assumed that the Muslim in- 
habitants had been settled in the city by order of the 
sultan (surgtin). By the beginning of the 10th/16th cen- 
tury, Selanik had about doubled in size, as apart from 
1,715 Muslim households, there were now 1,688 
Christians and 754 Jews. A high point was reached in 
925/1519, when an abbreviated register (idjmal) 
recorded 1,374 Muslim, 1,387 Christian and 3,143 
Jewish households. By about 967-8/1560, a significant 
drop in population had occurred (773 Muslim, 1,047 
Christian and 2,645 Jewish households). This decline 
was even more pronounced by 1022/1613, when the 
relevant figures were 1,090, 561, and 2,033, showing 
a relative increase of the Muslim and a decline of the 
Christian element. Thus Selanik seems to have held 
about 10,000 inhabitants in 883/1487 and to have 
oscillated between 18,000 and 30,000 thereafter. 
Ewliya Celebi claims 33,000 houses for llth/17th- 
century Selanik, which would give a population of 
over 150,000. But European travellers indicate that 
during its years of prosperity in the second half of the 
12th/18th century, Selanik possessed a population of 
about 60,000 to 70,000, 28,000 to 30,000 of whom 
were Turks. 

Among the revenue sources of Selanik and other 
towns of the area which the 9th/15th-century Ot- 
toman state attempted to exploit, were the salt pans, 
supplemented by a fishing weir in the vicinity of 
SelSnik itself. Accounts begin in 873/1468-9, but show 
that the enterprise was in constant difficulties. Several 

tax farmers were executed for their failure to pay, 
while the most prominent of them, a member of the 
Palaeologi family, managed to escape. The early tax 
farmers were Christians; but a Muslim miiltezim, 
recorded in 881/1476-7, equally owed a major sum. 
This situation indicates considerable disorganisation 
among sheepbreeders and fishermen, and possibly a 
crisis in the Macedonian village economy as a whole. 

Among the Ottoman monuments of the city, the 
most prominent is the White Tower, built probably in 
942/1535-6 upon the order of Sultan Siileyman; 
Mi c mar Sinan may have been responsible for its con- 
struction. Ewliya Celebi, who visited Selanik about 
1078-9/1668, has left a detailed description of the for- 
tifications, including the castle of Kalamarya; he was 
greatly impressed by their size, but felt that they had 
recently been neglected. Among the mosques that 
were built by Hafsa Khatun, the daughter of Hamza 
Beg, was a new building (872/1467-8, enlarged in 
1000-1/1592 and repaired in 1029/1620). This struc- 
ture had come to be known by the late 10th/ 16th cen- 
tury as the mosque of Hamza Beg. It is one of the very 
few mosques not founded by a sultan to possess a col- 
onnaded courtyard, albeit of irregular shape. The 
reconstruction around 1029/1620 is documented by a 
wak'if-ndme penned by the stylist Mehmed NergisT, 
while the donor was a kapudji by the name of Mehmed 
Beg son of Seyyid Ghazi. The second mosque con- 
structed at this time is known as the Aladja c Imaret; 
it was built by the former Grand Vizier Inegollii Ishak 
Pasha in 892/1486-7. It is of the T-shaped type, and 
its side chambers originally may have been meant to 
house dervishes. The mosque was associated with a 
public kitchen. Next to the building there once stood 
a minaret ornamented in multi-coloured stone, which 
gave the whole complex its name. Other major mos- 
ques were converted Christian churches; since Selanik 
had been taken by assault, the transformation of chur- 
ches into mosques was legally possible at any time. 
Many churches were taken over either around 905- 
6/1500, when refugees from Spain caused a wave of 
anti-Christian feeling, or in the tense years surround- 
ing the millenium of 1000/1591-2. The church of St 
Demetrius was converted into a mosque in 898/1492- 
3 and thenceforth known as the Kasimiyye, while in 
927/1521 local conflicts resulted in the confiscation of 
the Hagia Sophia, at that time the metropolitan 
church of Selanik. The funerary monument of 
Galerius, converted into a church in Byzantine times, 
was turned into a mosque on the initiative of the der- 
vish shaykh Khortadji. with the support of the Grand 
Vizier Sinan Pasha (998 or 999/1589-90 or 1590-1). 

In 883/1478, the Selanik tax register had not shown 
any Jewish inhabitants, probably because the city's 
Romaniote community had been transferred to Istan- 
bul by Sultan Mehemmed II (r. 848-50/1444-6 and 
855-86/1451-81). This move probably formed a part 
of the latter's programme of forced settlement (siirgun) 
to repopulate his new capital. But after the Jews had 
been driven out of Spain in 1492, and out of Spanish 
possessions in Italy shortly afterwards, Bayezld II in- 
vited them to settle in the Ottoman Empire. Many of 
these refugees were established in Selanik. A sum- 
mary (idpndt) register of 925/1519 records the 
presence of about 16,000 individuals; this figure in- 
cluded refugees from Germany and Italy as well. 

The Jewish immigrants continued to use Spanish in 
vernacular communication; when the language was 
committed to writing, the Hebrew alphabet was 
employed. A small number of Judaeo-Spanish 
speakers exists to the present day; but centuries' long 
isolation from Spain has resulted in many expressions 

archaic in standard Spanish. A number of Hebrew, 
Ottoman and Greek loanwords have also been incor- 
porated, and at one time there existed a second more 
"literary" version of Judaeo-Spanish, often called 
Ladino, which contained many expressions closely 
modelled on Hebrew. 

The new settlers engaged in the manufacture of 
woollen cloth, using water-powered fulling mills 
located outside the city. This manufacture of 
medium-quality cloth (cuka) supplied the market as 
well as the Janissaries. According to samples sent 
from Istanbul, the cloth was woven, dyed and fulled. 
Originally this service was paid for; but after the mid- 
dle of the 10th/16th century, the Salonica Jews were 
required to pay part of their taxes in the shape of 
woollen cloth. Their supply of raw wool was ensured 
by the privilege of purchasing whatever they needed 
from Balkan producers before any other customers 
could be supplied. A rabbinical regulation, pronounc- 
ed around 946-7/1540, threatened all Jews who ex- 
ported raw wool and the indigo needed for dyeing 
with excommuniction. However, the cloth manufac- 
ture, prosperous in the early 10th/16th century, ran 
into trouble thereafter. Raw wool prices increased, 
first because of Venetian demand, and when the 
Venetian woollen industry steeply declined after 
1008/1600, French purchasers prevented a fall in 
prices. Civilian demand for the finished product also 
fell away as English cloth appeared on the Ottoman 
market in large quantities after the 990s/ 1580s. Due to 
the need to supply the Janissaries, the manufacturers 
were not able to change their trades. In consequence, 
a large number of Salonica Jews emigrated in the 
11th/ 17th century; some went to other Rumelian 
towns, but the burgeoning Anatolian port of Izmir 

Nevertheless, at the same time, immigrants con- 
tinued to arrive, particularly from Livorno and other 
Italian commercial centres. Due to the contacts which 
these immigrants (Francos) brought with them, many 
of the most prosperous Jewish merchants of Selanik 
down to the 13th/ 19th century came from this group. 
But even though certain traders continued to be suc- 
cessful, on the whole Jewish merchants and financiers 
were eclipsed by other groups. An increasing orienta- 
tion of Balkan trade toward Europe facilitated the rise 
of Christian merchants, and as the Jewish traders of 
Selanik found fewer business opportunities, the com- 
munity in the 12th/18th century increasingly con- 
sisted of petty traders and artisans. 

In the 10th/16th century, the Jewish community of 
Selanik possessed a considerable scholastic activity, as 
local rabbis grappled with the religious and legal 
problems ensuing from life in a new environment. In 
addition, arrangements had to be devised to accom- 
modate those people who had been converted to 
Christianity in Spain or Portugal but wished to return 
to Judaism. Responses to legal questions (responsa) 
were at times committed to print, the first printing 
press in Selanik being established in 915-16/1510. Yet 
there were few contacts with either Muslim or Chris- 
tian scholars, and isolation in the long run resulted in 
a certain sclerosis of intellectual life. 

For the Jewish community of Selanik, the major 
event of the 11th/ 17th century was the movement of 
Shabbatay Sebi (1036-7-1087/1626-76 [q.v.]), a rabbi 
from Izmir who claimed to be the Messiah. He found 
adherents in Jewish communities all over the eastern 
Mediterranean and even in eastern Europe, but due 
to the concentration of Jews and institutions of Jewish 
learning in Selanik, his partisans in this city were of 
strategic significance. Complaints from rabbis unwill- 

ing to regard Shabbatay Sebi as the Messiah led to the 
involvement of the Ottoman authorities, and Sebi was 
assigned to forced residence in Gelibolu (1076- 
7/1666). But after continuing effervescence among 
Ottoman Jewish communities, Shabbatay Sebi was 
brought to the sultan's court in Edirne and offered the 
choice between death and conversion to Islam. He 
chose the latter (1077/1666). While some of his 
adherents returned to the established communities, 
others followed his example and were converted; this 
process continued over a number of years and led to 
the formation of a group known as the donme [q. v. ] 
(converts). Donme used Muslim names and followed 
Muslim ritual including the pilgrimage to Mecca; but 
down into the present century, they married only 
among themselves and had the basic features of rab- 
binic law taught to their children. Certain well-known 
rabbis of Selanlk who did not convert also retained 
sympathies for the movement of Shabbatay. 

In the 12th/18th century, Selanik's trade expanded 
after a period of relative stagnation, as the port 
became the centre of a lively import and export trade, 
particularly with France and various Italian states. 
Foreign consuls and vice-consuls became more 
numerous (at the beginning of the century, the French 
consul had been the only foreign representative), and 
an increasing number of non-Muslim merchants pur- 
chased the "protection" of foreign consuls, often ac- 
quiring more or less fictitious positions as translators. 
In local administration, one of the key personages was 
the principal customs farmer. Both the wealthy aghas 
of the area and European merchants maintained good 
relations with him, often lending money to a person 
of their confidence so that he could acquire the posi- 
tion. Another powerful figure was the commander of 
the local Janissaries, who controlled a force of about 
7,000 men. 

Among the goods exported, wheat occupied an im- 
portant place, even though this was mostly contra- 
band, aghas with cifiliks in the Macedonian coun- 
tryside supplying the exporters. Grain speculation 
was widespread, landholders holding back supplies 
until prices had increased or else making deals with 
exporting merchants. Grain riots were not unknown. 
Raw wool often went to France, while cotton and cot- 
ton thread gained in importance with the growing 
mechanisation of cotton- weaving in England, and also 
supplied looms in Germany. Tobacco and raisins 
were also exported. Imports consisted of manufac- 
tured goods, particularly French woollen fabrics, but 
also of Venetian silks, which all but monopolised the 
local market until, during the last quarter of the 
12th/ 18th century, silks from Lyons became impor- 
tant. With the increasing production of coffee on the 
Caribbean islands, lecafe des ties began to compete with 
Arabian coffee, and sugar brought in by European 
merchants with Egyptian sugar. 

For distribution, the pre-existing network of Balkan 
fairs was available, at which even rural consumers 
were able to purchase the cheaper imported goods. An 
active internal trade existed, linking Selanlk with Iz- 
mir, Egypt, Crete and the Aegean islands. Soap, 
linens and citrus fruits were brought to the city from 
these areas, while luxury goods often came from 
Istanbul. Greek merchants were particularly active on 
the overland route linking Selanlk to Vienna and the 
fair of Leipzig. Greek merchants, often based in 
Selanlk, were so active in the Hapsburg Empire that 
Maria Theresa and Joseph II took measures to limit 
their business, forcing rich Greeks established in the 
Hapsburg domains to concentrate on banking. But 
due to the foreign "protection" which many wealthy 
merchants had acquired and to the resulting tax eva- 

sion, the payment of taxes and dues was often a heavy 
load on the "ordinary" Greeks of Salonica, and the 
community owed large sums of money to wealthy 

The decline of Selanlk as an international port 
toward the end of the 12th/ 18th century was due part- 
ly to political conflict within the city, but also to inter- 
national conjunctures, as the revolutionary and 
Napoleonic wars interrupted established trade routes. 
The general downturn in the Ottoman economy 
during the last decades of the 12th/ 18th century prob- 
ably constituted a contributing factor. The abolition 
of the Janissary corps in 1241/June 1826 had political 
and economic repercussions on the local level; the 

of the victims of anti-Janissary repression. The disap- 
pearance of the Janissaries led to the final eclipse of 
woollen cloth manufacture by Jewish artisans, and ag- 
gravated pauperisation. 

In the course of the 13th/19th century, certain 
notables from the Jewish community developed an in- 
terest in the creation of new enterprises, particularly 
the processing of tobacco and later the manufacture of 
cigarettes. Due to the overall expansion of trade, /he 
need for commercial employees with training in ac- 
counting and foreign languages also made itself felt. 
In the eyes of the notables it was necessary to train 
both a literate blue-collar workforce and specialist 
white-collar employees. This meant a thorough 
restructuring of the established Talmud-Torah 
schools, which down to this period had imparted basic 
literacy to boys only. Reform of the school system 
became a major bone of contention in the struggle for 
control of the community between the established rab- 
binical elite and the lay notables, a struggle not with- 


other Ottc 

-t of the 

munitiesofthe 13th/19th century. After 1276-7/18 
the notables were able to muster 
Alliance Israelite Universale, an a 
France by which mainly francophone Jews supported 
French-language schooling among Jewish com- 
munities of the Mediterranean basin. The Alliance 
sponsored not only schools for both boys and girls, but 
also (no more than partially successful) efforts to ap- 
prentice children, in addition to a network of social 
organisations. In the early stages, the Alliance schools 
also organised the teaching of Turkish to their 
students, and in the early 20th century, when the an- 
nexation of Selanlk by Greece was increasingly 
viewed as a possibility, began to teach Modern Greek 
as well. The reports of the Alliance schoolteachers to 
their employers in Paris constitute a valuable source 
for the social history of late Ottoman Selanlk. 

At the beginning of the 20th century, Selanlk was 
one of the centres of the Committee for Union and 
Progress (Ittihdd we Terakki DJemHyyeti [?.».]). The 
backbone of this group were officers frustrated by the 
manner in which the war against Macedonian rebels 
was conducted— in 1903 there had been a series of at- 
tacks against public buildings in Salonica itself. 
Mustafa Kemal (the later Ataturk), a native of 
Salonica but at the time stationed in Syria, visited his 
home town and helped found a branch of the '•Olhmanli 
Hiimyyet DiemHyyeli. This organisation, based on small 
cells, under the leadership of the Salonica telegraph 
official Tal c at Bey, expanded rapidly among officers 
and bureaucrats in Macedonia. Contacts were estab- 
lished with the Ittihad we Terakki exile group in Paris, 
but the Salonica group retained its political and 
organisational independence and played a key role in 
the events which led to the restoration of the Ottoman 
Constitution in 1908. 

Around 1900, Selanlk possessed one of the largest 


is of factory labour in the Ottoman Em- 
pire. Much of this labour force, particularly in the 
tobacco industry, was female, and consisted of 
teenage girls put to work by their families in order to 
earn their dowries. But among the male labourers, 
organisations midway between guilds and trade 
unions began to appear at this time. Particularly 
among the minority group of Bulgarian labourers, 
socialist tendencies showed themselves. A local group 
was recognised by the Second International. In 1909 
Abraham Benaroya brought out a socialist paper in 
Ottoman (Ottoman Turkish title: c Amele Ghazetesi), 
Greek, Bulgarian and Judaeo- Spanish. While the Ot- 
toman and Greek versions soon had to be given up 
due to lack of reader interest, the Bulgarian and 
Judaeo-Spanish versions did attract readers. Official 
repression followed, including Benaroya's banish- 
ment to Bulgaria (1911). 

The Ottoman history of Selanlk ended with the 
First Balkan War, which began with an occupation of 
northern Albania by Montenegro, followed by an 
ultimatum on the part of several Balkan states against 
the Ottoman Empire (October 1912). While the 
Bulgarian army advanced as far as Catalca on the out- 
skirts of Istanbul, the Serbian and Greek armies 
entered Macedonia, with the Greeks occupying 
Selanik on 8 November 1912. The Ottoman govern- 
ment ceded Selanlk in a peace treaty with Greece in 
March 1914, to the great distress of the city's Jewish 
population, whose spokesmen had strongly favoured 
the continuance of Ottoman rule. Salonica was rapid- 
ly transformed into a Greek city, particularly through 
emigration of the Turkish-speaking population, the 
reconstruction following the great fire of 1917 and the 
settlement of large numbers of Anatolian Greeks 
entering the country as a result of the Turco-Greek 
population exchanges of 1923. 

Bibliography: See the articles on the city in EI' 
Q.R. Kramers) and tA (M. Tayyib Gokbilgin); 
most references given byKramers have not been 
repeated here. Mehmed c Ashik, Menazir ul-'-awalim, 
Staatsbibliothek Berlin, M. Or. Quart. 1344; 
Suleymaniye Genel Kitaphgi Halet Efendi no. 616, 
fols. 15b- 16a; Ewliya Celebi, Seyihat-nime , viii, 
Istanbul 1928, 142-70; I.S. Emmanuel, Hisloire de 
I'industrie des tissus des Israelites de Salonique, Paris 
1935; J. Nehama, Histoire des Israelites de Salonique, 1 
vols., Salonica-Paris 1935-78; F. Babinger, Ein 
Freibrief Mehmed II des Eroberers fur das Kloster Hagia 
Sophia zu Saloniki, in Byzantinische Zeitschrift, xliv 
(1951); M. Tayyib Gokbilgin, XV. -XVI. asirlarda 
Edirne ve Pasa Livdst, vakiflar, mulkler, mukataalar, 
Istanbul 1952, 150-6; Semavi Eyice, Yunanistan'da 
turk mimart eserlen, in Turkiyat Mecmuasi, xi (1954), 
157-82; N. Svoronos, Le commerce de Salonique au 
XVIII Steele, Paris 1956; Apostolos Bakalopoulos, A 
history of Thessaloniki, tr. T.F. Carney, Salonica 
1963; G.T. Dennis, The second Turkish Capture of 
Thessalonica, 1391-1394 or 1430, in BZ, lvii (1964), 
53-61; F. Babinger, Ein turkischer Stiftungsbrief des 
Nerkesi vom Jahre 102911620, in Aufsatze und 
Abhandlungen zur Geschichte Siidosteuropas und der 
Levante, ii, Munich 1966, 38-50; A. Bakalopoulos, 
Zur Frage der zweiten Einnahme Thessalonikis durch die 
Turken 1391-1392, in BZ, lxi (1968), 285-90; G. 
Scholem, Sabbatai Sevi, the mystical messiah 1626- 
1676, tr. R.J. Zwi Werblowsky, Princeton 1973; 
Melek Delilbasi, Selanik'in Venedik idaresine gecmesi ve 
Osmanh-Venedik savasi, in Belleten, xl/160 (1976), 
573-88; G. Haupt and P. Dumont, Osmanh im- 
paratorlugunda sosyalisl hareketler, tr. Tugrul Artunkal, 
Istanbul 1977; R. Kreutel, Ein Kirchenraub in 

Selamk, in WZKM, lxix (1977), 73-90; S.J. Shaw 
and Ezel Kural Shaw, History of the Ottoman empire 
and modern Turkey, ii, Reform, revolution, and republic: 
the rise of modern Turkey, 1808-1975, Cambridge 
1977, 263-7; P. Dumont, Social structure of the Jewish 
community of Salonica at the end of the nineteenth century, 
in Southeastern Europe - L 'Europe du Sud-est (Pitts- 
burgh), v/2 (1978), 33-72; B. Braude, International 
competition and domestic cloth in the Ottoman empire 
1500-1650: a study in undevelopment , in Review, ii/3 
(1979), 437-54; P. Dumont, Sources inedites pour 
I 'histoire du mouvement ouvrier et des courants socialistes 
dans V Empire Ottoman au debut du XX' si'ecle, in 
Tiirkiye'nin sosyal ve ekonomik tarihi (1079-1920), ed. 
Osman Okyar and Unal Nalbantoglu, Ankara 
1980, 383-96; idem, Une source pour I'etude des com- 
munautes juives de Turquie: les archives de I'Alliance 
Israelite Universelle, in Prilozi za Orientalnu Filologiju, 
xxxvi(1980), 75-106; E. Petropoulos, La presence ot- 
tomane a Salonique, Athens 1980; H. Lowry, Portrait 
of a city: the population and topography of Ottoman Selanik 
(Thessaloniki) in the year 1478, in Diptycha, ii (1981), 
254-93; Melek Delilbasi, Sultan II. Murad'in fetihten 
soma (29 Mart 1430) Selanik 'le izledigi politika ve bir 
hamam kiiabesi, in Tarih Arastirmalan Dergisi, xiv, no. 
25 (1982), 361-4; V. Demetriades, Torzoypayla zrjs 
&taaa\ovix-r)c. xa-ua rr)v trzoxqv Try; mvpxoxpomac., 
Salonica 1983; D. Nalpandis (ed.), Thessaloniki and 
its monuments, Thessaloniki 1985; S. Vryonis, The 
Ottoman conquest of Thessaloniki, in A. Bryer and H. 
Lowry (eds.), Continuity and change in late Byzantine 
and early Ottoman society, Birmingham-Washington 
D.C. 1986, 281-321; Delilbasi, Selanik ve Yanya'da 
Osmanh egemenliginin kurulmasi, in Belleten, cli, no. 
199 (April 1987), 75-106; Minna Rozen, Contest and 
rivalry in the Mediterranean maritime commerce in the first 
half of the eighteenth century: the Jews of Salonica and the 
European presence, in REJ, cxlvii (1988), 309-52; M. 
Kiel, A note on the exact date of construction of the White 
Tower of Thessaloniki, in idem, Studies on the Ottoman 
architecture of the Balkans, London 1990, no. VI; 
idem, Notes on the history of some Turkish monuments in 
Thessaloniki and their founders, in ibid, no. I; S.J. 
Shaw, The Jews of the Ottoman Empire and the Turkish 
Republic, New York 1991 (ample bibl.); Meropi 
Anastassiadou, Les inventaires apres dices de Salonique a 
la fin du XIX si'ecle: source pour I'etude d'une societe au 
seuil de la modernisation, in Turcica, xxv (1993), 97- 
136; Esther Benbassa and A. Rodrigue, Juifs des 
Balkans, espaces judeo-iberiques, X1V'-XX' si'ecles, Paris 
1993; G. Veinstein (ed.), Salonique, 1850-1918, La 
"ville des Juifs" et le reveil des Balkans, Paris 1993 
(incs. extensive bibl.); Maria Kabala, Eppatot <jto 
nXaioio tou tSSrjviiiOu upa-rau? Bi(3XioTfpa(pia, in Syn- 
chrona Themata, lii-liii (July-December 1994), 106- 
17; Yildiz Sertel, Annem Sabiha Sertel kimdi neleryazdi, 
Istanbul 1993, 15-79 (social life ofdonmesin Selanik 
and emigration to Istanbul); Delilbasi, Via Egnatia 
and Selanik (Thessaloniki) in the sixteenth century, in E. 
Zachariadou (ed.), Via Egnatia, Rethymon, in 
press. (Suraiya Faroqhi) 

SELANIKI, MustafA Efendi (d. ca. 1008?/ 
1600?), Ottoman official and historian. 

Almost nothing is known of his early life or family 
background, or when he was born and died, but he 
identified himself with Salonica [see selanik] and 
called himself Selaniklu and apparently reached old 
age. What is known stems almost entirely from his 
History (see below), in which he details his official ap- 
pointments, his presence at various military events 
during the reign of Siileyman Kanunl (e.g. during the 
Szigetvar campaign in Hungary of 1566) and his sue- 


cessors and his own views on affairs. Amongst the 
many official posts which he held were mukdta'-adjt of 
the Haramayn (till 988/1580); he was a dawdddr; he 
was secretary of the silahddrs and then of the Sipahls 
[q.v.] (till 996/1589); in 999/1591 the Grand Vizier 
Ferhad Pasha appointed him ruzndmedji [q.v.]; he 
became muhasebedji of Anatolia (1007/1599); and 
shortly thereafter he disappears from recorded 

The TaMkh-i Seldnikibegms with events of 971/1563 
and closes with the escape of the Voivode Kasim from 
custody in Shawwal 1008/May 1600, thus touching on 
four reigns up to that of Mehemmed III. It is more a 
diary of events which came to the writer's notice than 
a formal chronicle, the composition of which he might 
have intended to do later. It becomes progressively 
more detailed from the end of Murad Ill's reign 
(1003/1595). Rather than consulting other histories, 
Selaniki seems to have relied on his contacts with the 
leading men of state and on official documents from 
the Diwdn-i Humdyun and elsewhere for his informa- 
tion; as well as mentioning the viziers of the time, he 
mentions also the poet Baki [q. v.] and the Sheykh ul- 
Isldm Sun c Allah Efendi. Although the History is a 
prime source for the period, it does not seem to have 
been widely used or copied (yet over 25 ms. copies of 
'it today) until the early 12th/ If ' ~' 

it of c, 

, Katib 

Celebi and Na'Ima is quite different, but Solak-2 
clearly used it, without making acknowledgement. A 
feature of Selaniki's work is that he not only relates 
events but also includes criticisms of these events and 
of what he perceived as the general decline of the 
Ottoman state; the ups-and-downs of his own official 
career, with its frequent appointments and dismissals, 
may have affected his attitude here. 

The History was partially printed at Istanbul in 
1281/1864-5, but no complete, critical edition existed 
till that of Mehmet ipsirli, pp. LXXXV + 1,008, 
Istanbul 1989. 

Bibliography: Von Hammer, GOR, iii, 750, iv, 

168, 181, 185d, 243, 435; Babinger, GOW, 136-7; 

Ipsirli, Mustafa Selaniki and his History, in Tarih 

Enstitiisu Dergisi, ix (1978), 417-72; lA, art. Selaniki 

(Bekir Kiitukoglu). 

(Mehmet Ipsirli, shortened by the Editors) 

SELCUK [see aya solOk]. 

SELIM I, in official documents Sellmshah, 
nicknamed Yavuz or the Grim, ninth Ottoman 
sultan (reigned from 7 Safar 918/24 April 1512 to 8 
Shawwal 926/21 September 1520), conqueror of 
eastern Asia Minor, Syria and Egypt, and the first Ot- 
toman sultan entitled Khadim al-Haramayn al- 
Sharifayn or Servitor of Mecca and Medina. 

The struggle for the throne, 1509-13. 

To comprehend the circumstances and nature of 
the fierce struggle for the throne between BayezTd's 
three sons Korkud, Ahmed and Sellm, we have to 
keep in mind that Turco-Mongol peoples firmly 
believed that sovereignty was granted exclusively by 
God and no human arrangement could determine 
who is going to be next on the throne. 

In fact, when the throne became vacant, the prince 
who was able first to reach the capital city and take 
control of the treasury had the best chance to be 
recognised as ruler. So each of the sons of the reigning 
sultan tried to get the governorship nearest to Istan- 
bul. Although respected as an intellectual versed in 
Islamic law and the fine arts, Korkud was thought to 
be less apt for an Ottoman ruler. Described as just 
and generous in Ottoman sources, Ahmed was at the 
beginning the most popular of the princes, and the 

many of the great men of state, including the Grand 
Vizier C A1I and the Sfema 3 , wanted him to succeed. 
The youngest of the three princes, Selfm was born in 
875/1470-1 in Amasya from prince Bayezid (Bayezld 
II) and c Ayshe ('A'ishe), the daughter of the Dhu '1- 
Kadrid ruler c Ala> al-Dawla. When the rivalry for the 
throne began in 1509, Ahmed was governor of 
Amasya, nearest to Istanbul while Korkud was the 
governor of the distant sandjak of Antalya and Selfm 
that of Trebizond, the farthest of all. 

In 1509 Bayezld II [q.v.], an ailing old man, was 
believed incapable of leading the empire's armies 
against Shah Isma'il 1 [q.v.] of Persia, who had 
become a serious threat to the Ottomans, not only on 
the eastern frontiers of the empire but also within 
Anatolia through the activities there of his Turcoman 

In 1507, the Shah's invasion of the Dhu '1-Kadr 
[q.v.] principality, during which he passed over the 
Ottoman lands and enrolled in his army Turcomans 
who were Ottoman subjects, was considered a daring 
violation of Ottoman sovereignty. While Bayezid 
avoided any open conflict, Sellm from Trebizond took 
the initiative and in retaliation raided the Shah's ter- 
ritory as far as Bayburd and Erzincan. In Istanbul, 
this was interpreted as insubordination and caused the 
first rift between the sultan and his son. While, by his 
submissive attitude, Prince Ahmed was favoured by 
the sultan and the Grand Vizier, Sellm became the 
symbol of an aggressive policy. Sellm, however, 
declared that his concern was not to secure the throne 
but to save the empire from the havoc in which it had 
fallen. Openly criticising his father's inactivity, he 
showed himself as a champion of the warfare against 
heretics as well as Christians. Already from Trebi- 
zond he had organised raids into the neighbouring 
Georgia. His ghazd activities, used as political prop- 
aganda, won him the favour of the Janissaries, the 
timariot Sipahls and Akindjis [q. v. ] in Rumeli. It was 
the military campaigns that gave opportunity to these 
military classes to get promotion, more valuable timdrs 
or booty. In reality, for Selfm this was a struggle for 
survival since his and his son Siileyman's lives would 
be at stake should one of his brothers become sultan. 
Since there was little chance for him to reach Istanbul 
when the throne became vacant, he insisted that his 
governorship be exchanged for one in Rumeli. 

In this strategy, his first success was to secure the 
governorship of Kefe [q.v.] or Caffa for Suleyman 
(August 1509). When the news reached him that 
Bayezid was ill and was prepared to abdicate in favour 
of prince Ahmed, Korkud and Sellm suddenly left 
their seats, the former moving from Tekke to Manisa 
and the latter from Trebizond to Caffa. In fact, 
Bayezld favoured Ahmed as his successor, and openly 
expressed this at a meeting. From Caffa, Selfm in- 
sisted that he be appointed to a sandjak on the Danube, 
ostensibly to fight against the "unbelievers". When 
this was denied, he crossed the Danube at the head of 
about 3,000 men and marched toward Adrianople 
(March 1 5 1 1 ) . c All Pasha had the sultan declare Sellm 
a rebel, and Bayezid led an army of 15,000 to 
Adrianople, ordering at the same time all the 
Rumelian troops to join him there. At this juncture, 
one of the khalifes of Shah Isma'Tl in Tekke [q.v.], tak- 
ing advantage of the anarchical conditions in the em- 
pire, rose up and with his fanatical Kizflbash [q.v.[ 
followers and others, defeated the imperial troops sent 
against him (early March 1511). Korkud and Ahmed 
were held responsible for this critical situation, which 
strengthened further Selfm's position. Under these 
conditions, Bayezfd eventually yielded and agreed to 

give Selim the governorship of Semendere [q. v. ] or 
Smederovo. The sultan also vowed that "while he was 
alive he would not allow any of his sons to replace him 
in the sultanate and said that, at his death, it will be 
God's decision who would succeed him." Believing 
that he had pacified Selim, Bayezld returned to Istan- 
bul (24 August 1511). When Ahmed learned of the 
agreement between his father and Selim, he feared 
that Selim would become powerful enough to seize the 
throne, so he himself threatened to rebel in Anatolia. 
In Rumeli, Selim, trying to muster under his banner 
the troops of Rumeli at Eski-Zagra, heard that 
Bayezld and C A1T had actually decided to invite 
Ahmed to Istanbul and place him on the throne. But 
the Grand Vizier then had to cross in haste over to 
Anatolia, with 4,000 Janissaries (May 1511), in order 
to suppress Shah-Kulu, who was threatening to cap- 
ture Bursa. Confident in eliminating Shah-Kulu. the 
Grand Vizier and Ahmed deliberated how, after vic- 
tory, they would go to Istanbul and proclaim Ahmed 
sultan. Informed of this plan, Selim suddenly turned 
and occupied Adrianople (RaW I 917/June 1511) at 
the head of the Rumelian army of 30,000, acting there 
as sultan. This was open rebellion, which could not be 
tolerated, and at the head of an army of 40,000 men 
and artillery, Bayezld hastened to confront his 
rebellious son on the battlefield near Qorlu (8 
Djumada 1/3 August 1511). Selim, defeated, joined 
his son Suleymarl at Caffa. Almost at the same time, 
both C A1I Pasha and Shah-Kulu fell in a bloody com- 
bat in central Anatolia. With the death of the Grand 
Vizier, Ahmed lost his principal supporter for the 
sultanate. Ahmed now threatened to occupy Bursa 
with the Anatolian troops and to confront Selim. The 
empire was on the brink of a civil war. Ahmed's sup- 
porters pressed the sultan to invite him as soon as 
possible to Istanbul. However, the new Grand Vizier 
Hersek-oghlu Ahmed [q.v.] did not agree. Angry with 
Selim, the old sultan invited Ahmed to Istanbul, to 
march at the head of the army against Selim (26 
Djumada II 917/21 September 1511). But the 
Janissaries rebelled in favour of Selim, and the sultan 
had to yield, ordering Ahmed to return back to his 
sandj_ak. Now in open rebellion, Ahmed occupied the 
governorship of Karaman, where the anti-Ottoman 
Turcoman tribes were promising their support. Pro- 
Safawid Turcomans, now under the commander Nur- 
C A1I sent by the Shah (March 1512), rebelled in the 
Tokat area. Under pressure, Bayezld now decided to 
invite Selim to Istanbul (March 1512), since he was 
now considered by all as the only leader to cope with 
the critical situation. Confident of the support of the 
Kapi Kulu [?»] element of the troops, and of the 
Rumelian army, Selim was already on his way from 
Caffa to Istanbul. In the meantime, encouraged by 
Selim 's opponents Korkud arrived in Istanbul, hop- 
ing to ascend the throne with the support of the 
Janissaries, although the majority of these last were 
favouring Selim (early April 1512). In Istanbul, Selim 
was greeted by all dignitaries, including Korkud, on 
2 Safar 918/19 April 1512. The old sultan still had no 
intention to abdicate; but when Selim arrived at the 
court with a contingent of Janissaries, Bayezld was 
compelled to relinquish power. The deposed sultan, 
on his way to Dimetoka, died at the village of Abalar 
near Hafsa (25 Rabi* I 918/10 June 1512). The cause 
of his death was reported as suspicious by Menavino 
and Djennabl, but there is no hint confirming this in 
Venetian sources (von Hammer, GOR, iv, 86). 

Now, having received the formal bay'-a of the '■ulemd* 
and dignitaries and in control of the treasury and the 
Kapi Kulu troops, Selim became the only legitimate 

ruler of the empire. Ahmed, recognising the reality of 
Sellm's power, requested from him the governorship 
of Anatolia and actually began to appoint governors 
in his own name, ordering the soldiery to rally under 
his banner, and turning to the rebellious Turcoman 
tribes in the Tokat-Sivas area and in the Taurus 
mountains. Passing over to Anatolia at the head of his 
army, Selim expelled Ahmed's son 'Ala 5 al-DIn from 
Bursa (15 Djumada 918/29 July 1512) and moved to 
Ankara, from where his forces expelled Ahmed and 
his sons, who then fled to Shah Isma'Il for aid. 

Now, in order to be able to confront the Shah in a 
major campaign in the east, Selim had to eliminate in 
his rear all possible rivals for the sultanate. Hence he 
ordered the execution of all of the five sons of his de- 
ceased brothers between the ages of 7 and 21 who had 
taken refuge in Bursa. Next, at the head of 10,000 
men, Selim surprised his brother Korkud in his palace 
in Manisa, finally capturing and killing him. In the 
meantime, Ahmed had returned to Amasya, and in 
the winter of 1512-13, confronted Sellm's army on the 
plain of Yenisehir (27 Muharram 919/15 April 1513). 
Ahmed was defeated, captured and strangled. His son 
c Othman shared the same fate, while his other son 
Murad was with the Shah IsmacIl in Persia preparing 
to recover his father's patrimony. 

The campaign against Shah Isma c Il. 

Before he marched against Isma c Il, Selim had first 
to deal with the Kizilbash in his territory, who had 
already risen in the eastern provinces, while Selim 
was busy against his brother Ahmed. Ahmed's son 
Murad was ready to invade the area with the Shah's 
support. Selim conducted a purge of suspected 
Kizilbash, and 40,000 suspects were jailed or ex- 

Selim also took unusual measures for the period to 
deprive the Shah of the main cash revenue from the 
Persian silk trade. In the spring of 1514, he ordered 

the Ottoman lands and Europe. Later, he extended 
the embargo to include the Arab lands, which caused 
an additional friction with the Mamluks. He declared 
that any Persian, Arab or Turk found with Persian 
silk in his possession was subject to having his cargo 
seized, and in 1518, the sale of raw silk was altogether 
banned in Ottoman territory. 

On his way against the Shah, in Erzincan, the 
Janissaries began to mutter, but Selim did not hesitate 
to send to the executioner their mouthpiece, Hemdem 
Pasha, a governor. Shah Isma'Il was convinced that 
the Turcoman Kizilbash and the anti-Sellm governors 
of Anatolia would join him, hence he moved from the 
pasture lands of Tabriz to confront Selim at Caldiran 
[q.v.] in mid-August 1514. The two armies met at the 
plain there (2 Radjab 920/23 August 1514); in a 
furious assault with his forty thousand heavy cavalry, 
the Shah overpowered the c azebs, Ottoman light infan- 
try in the first line, and routed the Rumelian divisions 
on the left wing of the Ottoman army, then turning, 
attacked the centre where Selim was standing with his 
Janissaries. The stiff resistance of the Janissaries, 
decimating the Shah's cavalry with salvos of fire, and 
the war chariots tied with chains forming an im- 
pregnable stronghold, determined the outcome of the 
battle. Wounded by a bullet, the Shah barely escaped 
capture. His defeat has been attributed to a lack of 
firearms in his army (the earliest reference to his 
possession of muskets dates back to the year 1515: 
TKSA 6320; Tansel 88). After the victory, Sellm's 
plan was to pass the winter at Karabagh and to 
resume the campaign against the Shah next spring 
(for fath-names, see Ferldun, i, 386-9; Ibn Tulun, ii, 

47-53). But the insurgent Janissaries forced the sultan 
to return to Istanbul (for the Shah's embassies to 
Sellm after Caldiran and his attempts to find allies 
against the Ottomans, see Bacque-Grammont, 73- 
145). On 15 Radjab 920/5 September 1515, Selim 
entered Tabriz. After nine days, he left the city, tak- 
ing with him about one thousand citizens, artists, ar- 
tisans and rich merchants for Istanbul. On his way 
back to Amasya, where he spent the winter, Selim an- 
nexed the cities of Bayburd, Erzincan, Karahisar and 

One important consequence of the Ottoman victory 
was the Turkish conquest of all of the Shah's posses- 
sions in eastern Asia Minor, from Erzincan 
southwards to Diyarbakr and northern c Irak. The 
Kizilbash fortress of Kemah [see kemakh], a key 
stronghold on the crossroads of Erzincan and the 
Euphrates valley was taken by Sellm on 5 Rabi 11 I 
921/19 May 1515. The Shah's Kizilbash Turcoman 
governors and garrison commanders put up a stiff 
resistance to the Ottomans, while most of the Sunni 
Kurdish beys, who under the Kizilbash domination 
had lost their hereditary patrimonies, submitted. 
Through the activities of Idris Bidllsi [q.v.], a former 
Ak Koyunlu state secretary with close acquaintance of 
the Kurdish beys, Sellm followed a conciliatory policy 
to attach these Kurdish lords to his side, recognising 
with official diplomas their hereditary rights. Idris's 
list of the submitted Kurdish lords included those of 
Soran, c Imadiyye and Bukthi, who now began to at- 
tack the Persians and their allies. In the same regions, 
the Ak Koyunlu princes expelled by Isma c il also co- 
operated with the Ottomans. The population of 
Diyarbakr rose against the Shah, offering submission 
to Selim and appealing to him for aid, which he was 
only able to provide after the campaign against the 
Dhu '1-Kadrids (summer 1515) under the able com- 
mand of Biyikll Mehmed Pasha [q.v.], who entered 
Diyarbakr in Sha'ban 921/mid-September 1515. 
Following Biyikli's unsuccessful siege of Mardin 
[q.v.], the Persian forces came back and besieged 
Diyarbakr; Ottoman control over the Diyarbakr 
region was only achieved after Biyikli, reinforced with 
the army of Anatolia, had won a decisive victory over 
the Turcomans at Kargin-Dede. Thereupon, the for- 
tresses of Ergani, Sindjar, Cermik and Birecik sur- 
rendered, whilst Mardin, Hisn Kayfa, Ruha, Rakka 
and Mawsil fell later in 1516. 

The conquest of the Dhu '1-Kadrids. 

While advancing against the Shah in 1514, Sellm 's 
rear was threatened by the Mamluk sultan, who 
mustered forces in Aleppo, and by his vassal, the Dhu 
'1-Kadrid c Ala> al-Dawla. c Ala 3 al-Dawla, on Sultan 
Kansuh al-Ghawri's [q.v.] instructions, intercepted 
food supplies from his territory to the Ottoman army. 
C AU b. Shahsuwar, the son of the former ruler of the 
Dhu '1-Kadrid principality, now in Sellm' s service, 
began to invade his father's lands after Caldiran and 
with SelTm's support (winter 1514-15). Kansuh pro- 
tested against this as an infringement of Mamluk ter- 
ritory, but after the fall of Kemah, Selim decided to 
annex this Turcoman principality to his empire. 'Ala 3 
al-Dawla was killed by c Ali (29 Rabl c II 921/12 June 
1515), and Sellm sent his head to the Mamluk sultan. 
This strategically important region, inhabited by Tur- 
comans who had joined c Ali b. Shahsuwar during the 
Ottoman invasion, was left under his control. Ot- 
toman law was only imposed under Suleyman I, when 
Sipahis of Karaman were granted timdrs on the lands 
of the local Turcoman military elite. 

The campaign against the Mamluks. 

During Selim's campaign against the Persians, 

Kansuh had remained neutral, and Selim was careful 
on his part not to offend him. But after Caldiran, the 
Mamluk promised the Shah to attack Selim from the 
rear if he attacked the Shah again. Ottoman activities 
against the Dhu '1-Kadrids made hostilities 
unavoidable. Before war began, Selim took a series of 
measures to win over to his side the Arabs and some 
of the leading amirs, declaring that the Mamluks were 
a foreign caste of Circassians, dominating and op- 
pressing the great mass of Arab population. In fact, 
the Aleppo citizens promised to welcome the Ot- 
tomans in their city. The Syrian cities had become 
commercially dependent on the Bursa market, whilst 
it had become evident that the Mamluks were 
powerless to protect Arab merchants in their trade 
with India against the Portuguese, who now were in 
the Red Sea threatening to capture Mecca and 
Medina. In 1510 Kansuh himself had appealed to the 
sultan for aid to build a fleet at Suez, and Ottoman ex- 
perts and mercenaries [see rumi] were already in 
Suez, Djidda and Yemen. Kha'ir Bey, governor of 
Aleppo and Djanberdl Ghazali, governor of 
Damascus, both established secret relations with the 
Ottomans, and later, Sellm won over by promises of 
rewards many other Mamluks with promises of 
employment in the future Ottoman administration. 

The Mamluks feared that the Ottomans were going 
to invade Egypt from the sea. In fact, Ottoman ac- 

arsenals were intensified in 1515. Always declaring 
that his preparations were aimed at the heretic Shah. 
Selim claimed that by allying himself with Isma'Il, 
Kansuh was attempting to impede the Ottoman sultan 
in his efforts to extirpate heresy in Persia. The fatwa 
sought by Selim to legitimise his campaign against 
this Sunni Muslim ruler laid emphasis on Mamluk 
oppression and injustices committed against Muslims. 
Kansuh countered that the Ottoman sultan was using 
Christian soldiery in his army against Muslims. Upon 
Kansuh's formal demand for the evacuation of Dh u 
'1-Kadrid territory by the Ottomans, war was 
declared. Selim entered Mamluk territory in Malatya 
(end of July 1516), and the two armies confronted 
each other at the plain of Mardj Dabik [q. v. ] 40 km 
north of Aleppo on 25 Radjab 922/24 August 1516. 
Here, too, the Ottoman wagenburg tactics with the 
300 chained war chariots and their superiority in fire- 
arms determined the outcome of the battle. Kansuh 
was among the dead. Kha'ir Bey surrendered Aleppo 
and served the Ottomans faithfully, dying as Ottoman 
governor of Egypt in 1522. Selim left Aleppo after 
eighteen days, and reached Damascus on 1 Ramadan 
922/28 September 1516, where he spent the winter 
months. Although his viziers were not in favour of a 
campaign against Egypt, Selim was urged on by 
Kha'ir Bey and other Arab leaders against the newly- 
elected Mamluk sultan, Tumanbay, and he ordered 
preparations for the invasion of Egypt. An order was 
sent to Istanbul for the imperial fleet's departure for 
Egypt. The crucial problem was how to get the Ot- 
toman army through the Sinai desert to Egypt, and to 
provide a water supply, 30,000 water bags carried by 
15,000 camels were prepared. Declaring his decision 
to take all Muslim lands under his protection, Selim 
invited Tumanbay to recognise him as his suzerain; 
this was naturally refused. On their drive to Egypt, 
the Ottomans won their first victory near Ghazza 
against the Mamluk forces under Djanberdi (27 Dhu 
'1-Ka c da 922/21 December 1516). Leading Bedouin 
chiefs submitted to Selim. To confront the Ottoman 
army, Tunanbay had prepared a strong line of 
defence reinforced with artillery and ditches at al- 

Raydaniyya. On 29 Dhu '1-Hidjdja 922/22 January 
1517, while his main forces attacked in front, Selim 
surprised the Mamluks by circumventing Tuman- 
bay's fortified encampment. In the first hours of the 
combat, the vehement attack of the heavy Mamluk 
cavalry shook the Ottoman lines. But here, too, the 
outcome of the battle was determined by Ottoman 
superiority in fire-arms, foiling Mamluk cavalry at- 
tacks. The first Ottoman forces entered Cairo on 3 
Muharram 923/26 January 1517. Tumanbay and 

c able i 

cape i 

fighting in the streets of Cairo, refusing an offer of 
amdn \q.v.\ by Selim. Tumanbay mustered his troops 
on the west bank of the Nile, until Selim decided to 
cross the Nile and crush resistance. Tumanbay was 
captured and executed (15 April 1517). The Cairenes 
recognised Selim as their legitimate ruler, but only 
when he believed it was safe did he enter the city (23 
Muharram 923/15 February 1517). In the clashes in 
Cairo and outside the city, the number of Circassians 
killed or executed was estimated at 50,000 to 60,000 
(sec fath-ndmes in Feridun, i, 427-49; Ibn Tulun, ii, 

Following the fall of the Mamluk sultanate, the 
dependent Arab lands, including the Sharifs of Mecca 
and the Yemen, recognised Selim. 

Selim appointed Kha'ir Bey as Ottoman governor 
of Egypt, who succeeded in reconciling the remaining 
Mamluks and Arab shaykhs with the Ottoman ad- 
ministration. Before his departure from Cairo on 26 
Sha'ban 923/13 September 1517, Selim sent by sea to 
Istanbul 800 Cairenes who were thought "to cause 
trouble", including the last c Abbasid caliph al- 
Mutawakkil and many artisans. During long stays in 
Damascus and in Aleppo, he busied himself with 
organising Syria as a typical Ottoman province. He 
appointed governors and surveyors to register the 
population and revenues. He appointed Djanberdi 
governor of al-Sham (5 Safar 924/16 February 1518), 
and Nasir al-DIn Muhammad Ibn al-Hanash was 
given a sandj_ak along with various iktah; his control of 
Lebanon had been confirmed when Selim was on his 
way to fight against Tumanbay, but he later came in- 
to conflict with the Ottoman governors, leading to his 

Sellm's claim to pre-eminence in the Islamic 

The title of khalifa [see khilafat] was in common 
use among Muslim rulers when the Ottoman dynasty 
first emerged. It did not then represent dominion over 
the urnrna [q.v.] of all Muslims in the world, as was the 
case under the c Abbasids. However, the Mamluk 
sultans, taking under their protection Mecca and 
Medina and also the c Abbasid caliphs after 1258, 
claimed a primacy among Muslim rulers. Already 
after his victory at Mardj Dabik, Selim began to con- 
sider himself as the successor of the Mamluks, and 
assumed their title of Khddim al-Haramayn al-Sharifayn. 
The historian Ibn Tulun witnessed this at a Friday 
khutba in Damascus. Sellm's name was mentioned 
with the titles of al-Imam al- c ddil and of Sultan al- 
Haramayn al-Sharifayn. But no contemporary source 

ceremony, al-Mutawakkil officially transferred his 
caliphal rights to Selim. In contemporary Arab 
sources, al-Mutawakkil is always mentioned as al- 
Khalifa, Amir al-Mu^minin, and Selim only after him as 
Malik al-Rim (in 926/1519: Ibn Tulun, ii, 78, 91). Ac- 
cording to the tradition and the interpretation of the 
'■ulerruP, the Mamluk sultan or Selim himself could not 
claim to replace al-Mutawakkil because they did not 
descend from the Prophet's tribe Kuraysh. However, 

as successor to the last Mamluk sultan, Selim claimed 
primacy in the Islamic world; in a letter to the Shir- 
wanshah [q. v. ] he claimed that God had charged him 
to fight against heresy, to bring order to the true laws 
of Islam and to protect the Pilgrimage routes for 
Muslims (Feridun, i, 439-45). At a time when the 
Portuguese had entered the Red Sea and were 
threatening possibly to capture Mecca and Medina, 
the protection of Islam had become a crucial issue for 
all Muslims and the Arabs in particular. Presenting 
themselves as the "foremost of ghdzis" Selim, and his 
successor Siileyman, claimed to be the protectors of all 
Muslims in the world. In 1526, the latter used the title 
wdrith al-khildfa al-kubrd "inheritor of the supreme 

At the time when developments in the east kept him 
busy, Selim was careful to maintain peace with 
Christian nations in the west, in particular with 
Venice and Hungary. On 17 September 1517 he 
renewed the Venetian capitulations in Cairo, with the 
additional stipulation that the Venetian tribute to the 
Mamluks of 8,000 gold ducats for Cyprus was to be 
paid thereafter to the Ottoman sultan. Sellm's 
diplomacy toward Christian nations was altogether 
successful; but Pope Leo X's increased efforts to 
organise a European crusade, and the Shah's and 
Kansuh's diplomatic relations with Western states 
against the Ottomans, did not in the end result in any 
hostile activity (K.M. Setton, The Papacy and the Le- 
vant, ii; Bacque-Grammont, 135-45, 168). 

The outbreak of a new Kizilbash Turcoman 
rebellion in the Amasya province in the spring of 1519 
under a dervish called Djelal, declaring himself Shah 
Well and a Mahdi [q.v.], showed that Shah Isma'fl 
was still the principal threat to the empire. The 
rebellion was suppressed with difficulty (April 1519). 
At the time of his accession to the throne, Selim was 
advised to make the conquest of the islands of Rhodes 
and Chios one of his most urgent tasks. In 1519 e: 


n of 

one hundred galleys, were believed in Venice to be the 
signal of a campaign against Rhodes (von Hammer, 
iv, 247-9). But the following year, Selim died on his 
way from Istanbul to Edirne near Corlu on 8 Shawwal 
926/21 September 1520. His only son Suleyman [q.v.] 
succeeded him without difficulty. 

Well-educated, the author of a collection of poems 
in Persian, an admirer of Muhyl al-DIn Ibn al- c Arabi 
[q.v.], Selim was at the same time an uncompromising 
autocrat, and a quick-tempered, merciless man. 

Bibliography: 1. Archives. TKSA: Topkapi 
Sarayi Arsivi, nos. E. 6532, 5594, 5970, 6382, 
3703, 6446, 3602, 3197, 5876, 12216, 6335, 6701, 
6118, 6189, 7055, 7084, 6631, 6399, 6193, 3192, 
6304, 5460, 5960, 6401, 12077, 5465, 6672, 5674, 
5469, 4796, 5293, 3295, 6384, 2261, 4467, 11634, 
8277, 5824, 4312, 6341, 6608, 5594, 6341, 10292, 
6401, 5460, 7082, 5846, 9658, 5674, 6420, 4744, 
8327, 6577, 5882, 9659, 8339, 6315, 2667, 6301, 
10149, 6396, 7053, 6927, 6532, 5811, 6185, 12363, 
7633, 6586, 5447, 9525, 6186, 3058, 6221, 7072, 
8312, 6425, 6522, 7292, 6316, 5452, 6333, 3057, 
6376, 6205, 5691, 7052, 2779, 12277, 6510; 
Osmanh Arsivi: Ali Emiri Tasnifi, 

1. Selim 



of i 

., Munshe'dl al-seldtin, Istanbul 1247, ii, 367- 
500; Sari c Abd Allah, DustUr al-inshd\ SOleymaniye 
Library, Esad EC no. 3333; Tadji-zade Dja'fer, 
Munshe^dt, Mevlana Miizesi Library, Konya no. 
4935; Halil Edhem, Mislr fethi mukaddemdttna Wid 
muhimm bir welhika, in TTEM, xix, no. 96 (1928), 

2. Chronicles. Hayder Celebi, Ruzndme, in 
Feridun, Munshe^at, 458-500, is an official journal of 
Sellm's reign. For the Selim-names of Ishak, Shiikrl, 
ShlrazI Keshfl, Shlrl, Muhyl, Sa c dl, SudjudI, 
Djewri, Djelal-zade Mustafa, Djelal-zade Salih, 
c ArifI, Sa c d al-Din, Idrls Abu '1-Fadl, c Abd al- 
Kablr, Lahml, Thanayl, Hayati, and Shuhudi, see 
Sehabeddin Tekindag, Selimndmaler , in Tarih 
Enstitiisu Dergisi, i (1970), 197-231, and Ahmet 
Ugur, The reign of Sultan Selim I in the light of the Selim- 
ndme literature, Berlin 1985; ismet Parmaksizoglu, 
Uskublii Ishak Qelebi ve Selimnamesi, in Tarih Dergisi, 
iii (1951-52); General: IdrTs Bidllst, Hasht bihisht, 
TKS Library 1540; for Ibn Kemal (Kemal Pasha- 
zade), Tawdrikh-i dl-i <0thman, defter IX, see Ahmet 
Ugur, above; Hadldl, Tawarikh-i al-i 'Othmdn, ms. 
B.L. Or. 12896; Silahshur (?), Fethndme-yi diydr-i 
^Arab, ed. Tansel, in TV, no. 17, 294-320; no. 18, 
430-54; Mustafa C A1T, Kunh al-akhbar, Suleymaniye 
Library, Es'ad Efendi no. 2162; Sa c d al-Din, Tddj 
al-tawdnkh, Istanbul 1280, 135-619; Rustem Pasha, 
Tdnkh, B.N. Paris, suppl. turc 1021; P. Horn, 
Diwdn-i YawGz Sultan Sallm, Berlin 1908; Hasan 
Rumla, Ahsan al-lawankh, ed. and tr. C.N. Seddon, 
Baroda 1934; Bidllsl, Scheref-ndmeh or histoire des 
Kourdes, ed. W. Weliaminof-Zernov, St. Peters- 
burg, 2 vols, 1860-2; Djar Allah al-Makkl, al- 
Djewdhir al-hisdn ft mandkib al-Sultan Suleyman ibn 
<Othman, Dar al-MathnawI, no. 360; Ahmad b. 
Zunbul, Ghazawdt al-Sultan Salim Khan ma 1 al-Sultan 
al-Ghawn, Cairo 1278, Tkish. tr. Ahmad Suhayll, 
Tdrikh-i Misr al-Djedid, Istanbul 1142; Ibn Iyas, 
BaddV al-zuhur fi wakdH c al-duhur, ed. Mohamed 
Mostafa, Cairo 1960; Shams al-DTn Muhammad 
Ibn Tulun, Mufdkahdt al-khilldnfi hawddith al-zamdn, 
1. Muhammad Mustafa, Cairo 1952. 


. Mar 

3. Eui 

Eugenio Alberi, Le relazioni degli ambasciatori al Senato 
durante il secolo decimosesto, iii/3, Florence 1855; J. 
von Hammer, GOR, ii, Pest 1828, 376-541, Tkish. 
tr. M. c Ata, Dewlet-i ^Othmdniyye tdnkhi, iv, Istanbul 
1330, 95-256; Francesco Sansovino, Gl'annali tur- 
cheschi, Venice 1573; gagatay Ulucay, Yavuz Sultan 
Selim nasd padisah oldu, in TD, vi/9 (1946), 53-90, 
117-42, viii/11-12 (1995), 185-200; Marie Therese 
Speiser, Das Selimndme des Sa c di b. Abd iil-Mute'dl, 
Ubersetzung, Zurich 1946; Elke Eberhard, Osmanische 
Polemik gegen die Safawiden im 16. Jahrhundert nach 
arabischen Handschriften , Freiburg i. Breisgau, 1970; 
Irene Melikoff, Le probleme ktzilbash, in Turcica, vi 
(1975), 49-67; Hanna Sohrweide, Der Sieg der 
Safaviden in Persien und seine Ruckwirkungen auf die 
Schiken Anatoliens in 16 Jahrhundert, in hi. , lx (1965), 
95-223; Faruk Siimer, Safavi devletinin kurulusu ve 
gelismesinde Anadolu turklerinin rolix, Ankara 1976; 
Adel Allouche, The origin and development of the 
Ottoman-Safavid conflict, 906-962/1500-1555, Berlin 
1983; §ehabeddin Tekindag, Yeni kaynak ve 
vesikalann isigi altinda Yavuz Sultan Selim 'in Iran seferi, 
in TD, xiii/22 (1968), 49-78; H. Jansky, Die 
Eroberung Syriens durch Sultan Selim I, in MOG, ii/3-4 
(1926), 173-241; idem, Die Chronik des Ibn Tulin ah 
Geschichtsquelle iiber den Feldzug Sultan Selims I. gegen die 
Mamluken, Vienna 1929; CeliaJ. Kerslake, The cor- 
respondence between Selim I and Kdnsuh al-Gawri, in 
Prilozija Orientalnu Filologiju, xxx (Sarajevo 1980), 
219-34; Palmira Brummet, Ottoman seapower and 
Levantine diplomacy in the age of discovery, New York 
1994; Irene Beldiceanu-Steinherr, Le regne de Selim 
I", in Turcica, vi, 34-48; Selahattin Tansel, Yavuz 
Sultan Selim, Ankara 1969; J.-L. Bacque- 

Grammont, Les Ottomans, les Safavides et leurs voisins, 
Istanbul 1987; Halil inalcik, Ottoman methods of con- 
quest, in SI, iii (1954), 103-29; idem, Bursa and the 
commerce of the Levant, mJESHO, iii (1960), 131-47; 
idem, The heyday and decline of the Ottoman Empire, in 
Cambridge hist, of Islam, Cambridge 1970; idem 
(ed.), An economic and social history of the Ottoman Em- 
pire, Cambridge 1994; 6.L. Barkan, XV. ve XVI. 

hukuki ve mali esaslart. I. Kanunlar, Istanbul 1943; B. 
Lewis, A Jewish source on Damascus just after the Ot- 
toman conquest, in BSOS, x (1940), 180-4; C.H. 
Becker, Barthold's Studien iiber Kalifund Sultan, in Isl. , 
vi, 386-412; W. Hinz, Das Steuerwesen Ostanatoliens 
im 15 und. 16. Jhdt. , in ZDMG, c (1950), 177-201. 



SELIM II, the eleventh Ottoman sultan (r. 974- 
82/1566-74), the third son and the fourth of the six 
children of KanunI Suleyman I and Khurrem Sultan 
[q.vv.]. He was born in Istanbul on 26 Radjab 930/30 
May 1524, during the festivities accompanying the 
marriage of Siileyman's Grand Vizier Ibrahim Pasha 
[q.v.]. Together with his elder half-brother Mustafa 
and his elder brother Mehmed, Selim was one of the 
three princes in whose honour was held the siinnet 
dugunii (circumcision feast) of 1530, one of the major 
dynastic spectacles of Siileyman's reign. He remained 
in Istanbul until appointed in 1542, at the age of 18, 
to his first provincial post in Konya as sandjak begi of 
Karaman. In 951/1544, following his brother Meh- 
med 's death, Selim was transferred to the latter' s 
more prestigious sandjak of Sarukhan [q. v. ] at Manisa, 
remaining there until his transfer back to Konya in 
1558. In 955/1548 Selim was temporarily assigned to 
Edirne to guard the European front, whilst Suleyman 
was on campaign in the east against Safawid Persia. 

Following the deaths in 960/1553 of Mustafa and 
another brother Djihanglr, Selim and his younger 
brother Bayezld were the only surviving sons of 
Suleyman. On the death of their mother Khurrem in 
965/1558, rivalry between Selim and Bayezld broke 
out into an open succession struggle. With the aid of 
troops sent by Suleyman and led by the third vizier 
Sokollu Mehmed Pasha [q.v.], Selim defeated 
Bayezld's provincial forces at the battle of Konya in 
966/1559, forcing Bayezld to take refuge in Persia. 
After lengthy negotiations between Suleyman and 
Shah Tahmasp I [q.v.], Bayezld and his sons were 
assassinated at Siileyman's bidding in 969/1562 (S. 
Turan, Kanuni'nin oglu sehzade Bayezid vak'asi, Ankara 
1961, passim). Now Siileyman's sole heir, Selim was 
transferred to the sandjak of Kutahya, where he re- 
mained until Siileyman's death at Szigetvar in 
974/September 1566 whilst on campaign in Hungary. 

The Grand Vizier Sokollu Mehmed Pasha kept 
Siileyman's death secret for several weeks, enabling 
Selim to be safely enthroned in Istanbul after a hur- 
ried, secret journey from Kutahya. Selim then pro- 
ceeded to Belgrade to be acclaimed by the army and 
to escort Siileyman's bier. His reign began in confu- 
sion. Initially refusing to pay accession donatives at 
the level demanded by the Janissaries and household 
troops, Selim was forced to do so by rioting on his 
return from Belgrade to Istanbul. He was also obliged 
to grant timdrs or other awards to ca. 8,000 provincial 
troops recruited to his side in the 1559 fight against 
Bayezid. Thereafter, he took little part in govern- 
ment, retiring to the harem and delegating virtually all 
responsibility to Sokollu Mehmed Pasha, who re- 
mained Grand Vizier throughout his reign. 

During Sellm's eight-year sultanate, naval ac- 
tivities took precedence over land campaigns, and ac- 


tion in the extreme north and south of the empire 
replaced the prominence given in Siileyman's reign to 
east-west, Safawid-Habsburg campaigns. Relations 
with Persia remained subdued in the wake of the 

was signed in 1568 with the Austrian Habsburgs, 
stipulating an annual payment by the latter of 30,000 
ducats in respect of those parts of north-west Hungary 
claimed by the Ottomans and still under Habsburg 

In the eastern Mediterranean, the island of Sakiz 
[q.v.] (Chios) was captured from the Genoese by the 
kapudan pasha Piyale Pasha [q.v.] in 1566 (technically 
while Suleyman was still sultan, but notified to Sellm 
at the time of his accession), and Kibris (Cyprus) from 
the Venetians by Lala Mustafa Pasha [q.v.] in 1570-1, 
thus increasing the safety of Ottoman sea routes to 
Egypt. Together with proximate areas on the 
mainland of Anatolia, Cyprus was formed into a new 
beglerbegilik (province) and received a large influx of 
Turkish settlers. In the ensuing naval battle off 
Inebakhti (Lepanto) the Ottoman fleet was defeated 
(979/October 1571) by a Papal-Venetian-Spanish 
fleet commanded by Don Juan of Austria, but was 
rebuilt rapidly over the winter of 1571-2 and Ottoman 
naval supremacy in the area restored. In the western 
Mediterranean, the fortress at Tunis (Khalku '1- 
wa c ad, or Goletta) was lost to Spain in late 1572, but 
recaptured by Kodja Sinan Pasha [q.v.] in 982/1574 
and Tunis and its dependencies formally established 
as a new beglerbegilik. Elsewhere, an Ottoman fleet was 
sent from the Red Sea in 1568 at the request of the 
Muslim ruler of Sumatra to aid him against the Por- 
tuguese in the Indian Ocean, but this achieved little. 

On land, successful military operations were con- 
ducted against Arab tribal revolts in the Basra region 
(1567), and against Zaydi threats to Ottoman control 
in Yemen, which culminated in the loss of $an c a> 
(1567). During 1568-70, under the command first of 
Ozdemir-oghlu c Othman Pasha [q. v. ] and then of the 
governor of Egypt, Kodja Sinan Pasha, Yemen was in 
effect reconquered and set up as a single province, 
rather than two as previously. 

The most ambitious project of Selim's reign, the 
building of a canal between the rivers Don and Volga 
(attempted 1569-70), was unsuccessful. It was pro- 
moted by Sokollu Mehmed Pasha with the three-fold 
object of protecting the pilgrimage route from Central 
Asia, of curtailing the southward advance of Muscovy 
(which had captured Kazan in 1552 and Astrakhan in 
1556), and of establishing the potential to attack Per- 
sia from the north. It would also have served to extend 
Ottoman control over the khans of the Crimea. 
Adverse weather conditions, unrest amongst the 
troops involved, and over-extended lines of communi- 
cation led to the abandonment of the project with only 
a third of the canal excavated (cf. H. inalcik, The 
origin of the Ottoman-Russian rivalry and the Don- Volga 
canal (1569), in Annales de I'Unwersite d'Ankara, i [1946- 
7], 47-110). 

Sellm II died aged 50 on 28 Sha'ban 982/13 
December 1574 in Istanbul (the first sultan to die 
there) following a fall in the palace hammam, and was 
succeeded by Murad III (982-1003/1574-95 [q.v.]), 
his eldest son by his Venetian khdssekiNur Banu [q. v.]. 
Five younger sons were executed and buried with him 
in his tomb in the courtyard of the Aya Sofya mosque. 
Three daughters were married to prominent viziers in 
a triple wedding in 1562: ismikhan to Sokollu Meh- 
med Pasha, Gewherkhan to Piyale Pasha, and Shah 
to Hasan Pasha (re-married later to Dal Mahmud 
Pasha). This network of dimdd (son-in-law) connec- 

tions, and Selim's reliance upon Nur Banu and his 
sister Mihrimah, encouraged the growth of the much- 
maligned "harem politics" (cf. L.P. Peirce, The im- 
perial harem: women and sovereignty in the Ottoman empire, 
Oxford 1993, passim). 

Known to Ottomans as Sari Sellm, "Sellm the 
Sallow", and to Europeans as "Sellm the Sot" 
because of his love of wine, Selim was skilled in ar- 
chery and particularly fond of hunting, for which he 
spent much time in Edirne. He was also an accom- 
plished poet, under the makhlas Sellmi (although no 
diwan of his survives), and a discriminating patron of, 
amongst others, the historian Mustafa C A1I, the poet 
Bakl [q.v.], and the shahnameaji Lokman (see e.g. F. 
Cagman, Sahname-i Selim Han ve minyaturleri, in Sanat 
tarihi yilltgi, v [1972-3], 411-43). He was the first Ot- 
toman sultan to honour members of the 'ulema'' with 
accession donatives. Among his major architectural 
projects were the repair of the Mecca water supply 
system and the re- roofing of the great mosque, the ad- 
dition of two minarets and extra buttresses to the Aya 
Sofya mosque, and the construction of the Sellmiyye 
complex by the architect Sinan [q.v.] in Edirne. 

Bibliography: 1. Contemporary Ottoman 
sources: Djelal-zade, ed. P. Kappert, Geschichle 
Sultan Suleyman Kanunls von 1520 bis 1557 ... von 
Celalzade Mustafa, Wiesbaden 1981, 117a, 194a ff., 
386b; Selanikl, ed. M. ipsirli, Tarih-i Seldniki, i, 
Istanbul 1989, 39-99, 101; Mustafa C A1I, Kunhii 7- 
dkhbdr, unpubl. mss. summarised in J. Schmidt, 
Pure water for thirsty Muslims: a study of Mustafa '■All of 
Gallipoli's Kunhii 'l-ahbar, Leiden 1991, 327-30, and 

Topkapi Sarayi Muzesi, Revan 1290; Pecewi! 
TaMkh, Istanbul 1283, i, 438-504. 

2. Further Ottoman sources, published 

those cited in the text are given in S. Turan, tA, art. 
Selim II. See also C.H. Fleischer, Bureaucrat and intel- 
lectual in the Ottoman empire: the historian Mustafa Ali 
(1541-1600), Princeton 1986, esp. 33-9; M.T. 
G6kbilgin, tA, art. Mehmed Pasa, Sokollu; J.-L. 
Bacque-Grammont, in R. Mantran (ed.), Histoire de 
I 'empire Ottoman, Paris 1989, 155-8 and also index. 
For additional bibl., see S.J. Shaw, History of the Ot- 
toman empire and modern Turkey, Cambridge 1976, i, 
318-19. (Christine Woodhead) 

SELIM III, the twenty-eighth sultan of the Ot- 
toman empire (1203-22/1789-1807), first son of 
Mustafa III and grandson of Ahmed III [q.vv.], was 
born in Istanbul on 27 Djumada I 1 175/24 December 

g the , 

. His 

father's more liberal outlook allowed Sellm con- 
siderable freedom of aetion, including the observation 
of the training of Baron de Tott's new Ottoman rapid- 
fire artillery corps, which was sent to the Danube 
battlefield in the last year of the 1768-74 Russo- 
Turkish war. Selim's childhood entourage included 
many individuals such as Abu Bakr Ratib Efendi and 
Kiicuk Husayn Pasha [q.v.] who would serve him and 
influence his thinking upon his accession. An abortive 
coup to replace Sultan c Abd al-Hamid I [q.v.] with 
Sellm, reputedly plotted by the Grand Vizier Khalll 
Hamid Pasha in 1785, forced the Sultan to restrict the 
young prince's movements, but that did not prevent 
Selim from maintaining contaet with the world 
beyond the palace. His culpability in the events of 
1785 has never been effectively demonstrated. The 
French ambassadors Saint-Priest and Choiseul- 

Gouffier encouraged Sellm to correspond with Loui: 
XVI, letters which demonstrate his powerful and 
Russian sentiments and wish for revenge. Choiseul 
Gouffier supported the sending of Ishak Bey, < 


effort t. 

rage the French link with the Ottoman heir- 

On 11 Radjab 1203/7 April 1789, at the age of 27, 
Sellm III was proclaimed Sultan, ascending the 
throne at one of the most difficult moments in the 
history of the dynasty, succeeding to an empire at war 
with both Russia and Austria and riven by internal 
rebellions. The previous autumn, the Austrians had 
made deep inroads into Ottoman territory, capturing 
Khotin [q.v.] on 9 September and routing the Ot- 
toman army at Slatina later that same month. In spite 
of the evidence of Ottoman military exhaustion, Sellm 
vigorously supported the war effort, immediately con- 
firming Khodja Yusuf Pasha in his position as Grand 
Vizier and Commander-in-Chief of the battlefront, 
glorifying past Ottoman successes, reinstating the ac- 
cession bonus to the Janissaries [see yeni-Ceri], which 
had lapsed with c Abd al-Hamld I, and generally 
boasting the morale of his people (Enwerl; Shaw, 32). 

Khodja Yusuf, the bellicose initiator of the war, had 
demonstrated some success in maintaining the Ot- 
toman position on the southern shores of the Danube, 
and in winning the loyalty of his soldiers, in spite of 
the Ottoman losses. He was opposed by Djeza'irli 
Ghazi Hasan Pasha [q.v.], sole hero of the 1770 
Ceshme [q.v.] naval disaster, reformer, critic of the 
war, and Grand Admiral of the Navy upon Selim's 
accession. Sellm replaced the experienced admiral 
Hasan with his boyhood friend Kiicuk Husayn Pasha, 
ordering Hasan to command the fortress of Ism5 c Il 
[q.v.] in an effort to recapture Ozii [see ozi], placating 
the advocates for continuing the war, consolidating 
his own power base, and managing to maintain the 
services of a valued commander. The new campaign 
season proved a disaster, however, culminating in the 
battle of Martineshti on 22 September 1 789 against 
combined Austrian and Russian forces, a total rout 
for the Ottomans. Thereafter, the Austrians occupied 
Belgrade [q.v.] and Bucharest [see bukresh], and the 
Russians Ak Kirman [q.v.] and Bender [q.v.]. 

An Ottoman treaty with Sweden to distract Russia, 
which was concluded on 11 July 1789, resulted in lit- 
tle, but an alliance with Prussia dating from 31 
January 1790, and the death of Joseph II of Austria 
in February, led to the conclusion of an Ottoman- 
Austrian treaty mediated by Prussia, England and 
The Netherlands, at Zistowa, on 4 August 1791, but 
only after Prussia and Austria had settled their dif- 
ferences in the Convention of Reichenbach the 
previous summer. By the stipulations of the Zistowa 
treaty, the Ottomans retained the territories in 
Wallachia [see eflak] and Moldavia [see boghdan] 
which had been occupied by Austria, essentially a 
recapitulation of the 1 739 Treaty of Belgrade, ceding 
only Old Orsowa as part of a new definition of the 
Austro-Ottoman border. 

Hostilities continued with the Russians, with 
disastrous results for the Ottoman forces, Russia oc- 
cupying many of the important Danube fortresses, 
notably Ismail, the Ottoman base of operations, after 
a long struggle on 22 December 1 790. After the winter 
hiatus, the Russians resoundingly beat the Ottoman 
forces south of the Danube at Macln in April 1791, 
the fortress itself capitulating on 9 July. Sellm was ill- 
served by his commanders, especially after the death 
of the newly-reappointed Grand Vizier Djeza'irli 
Ghazi Hasan Pasha on the battlefield in March of 
1790. His successor, Sherif Hasan Pasha, after allow- 

the Russians to overrun Budjak [q.v.], was ex- 
ed at Shumla in February, 1791, and replaced 
: again by Khodja Yusuf Pasha, who could no 
jer contain the complete breakdown of Ottoman 
defences. A truce was arranged by mid-summer, and 
■wed by the treaty of Jassy [see yash], 9 January 
I, recapitulating most of the articles of the 1774 
Kiicuk Kaynardja [q.v.] treaty, but establishing the 
lew Ottoman-Russian border at the Dniester and the 
Kuban rivers and ending conclusively Ottoman 
lis to the Crimea [see k!r!m]. 
elim faced a demoralised, practically bankrupt 
and highly decentralised empire, conditions exacer- 
bated by the costly and fruitless campaigns. 
Rebellions in Arabia, the Balkans and the Caucasus 
continued while he undertook to reform his ad- 
ministration, pursuing an energetic programme 
designed and carried through by his advisors. His in- 
tentions were clear from the early days of his succes- 
sion, when he summoned a council of 200 men of state 
in May 1789 to discuss the condition of the empire. 
(Djewdet, 2 vi, 6-7; Shaw, 73). During the war, Sellm 
pursued the re-ordering of the Ottoman military 
machine as best he could, but with little practical 
results. Immediately following the peace at Jassy, the 
Grand Vizier was ordered to solicit a number of writ- 
ten reform proposals. By far the most comprehensive 
was that of Abu Bakr Ratib Efendi, Ambassador to 
Vienna following the Zistowa peace treaty, whose 
analysis of Austrian institutions served as a recipe for 
the Ottoman reform programme, Nizam-i Diedid [q. v. ] 
or "New Order," the same term used to refer to the 

Addressing military reform meant facing powerful 
opposition, as the Janissary privileges in the form of 
pay and rations (esame) were deeply entrenched in all 
Is of Ottoman society. Significant reform was 
undertaken after 1793 in the artillery corps, with the 
introduction of new troops, new schools and training 
by foreign officers. Selim's energy, however, also ex- 
tended to the traditional Janissary and Sipahi [q.v.] 
corps, instituting discipline, introducing new 
weaponry, and undertaking to see that they were 
regularly paid and comfortable in rebuilt barracks in 
Istanbul, but to no avail. Establishing the Nizam-i 
Djedtd army proved more productive, an outgrowth of 
the irregular Lewend [q.v.] organisation, with 
separate barracks, the Lewend Ciftelik, located 
outside the centre of Istanbul, and training and 
discipline styled along western lines. Sellm felt confi- 
dent enough to issue regulations concerning the new 
corps only in 1794, a single regiment of 1,602 officers 
and men, attached to the old imperial BostandjT corps 
[q.v.], as its riflemen branch. By 1800, as a result of 
the threat posed by the Napoleonic invasion of Egypt, 
a second regiment was installed in the new Sellmiyye 
barracks in Uskiidar, a force drawn largely from the 
unemployed of the streets of Istanbul, and Turkish 
peasants from Anatolian villages, whose enlistment 
was encouraged by high salaries and tax exemptions. 
By 1801, the force numbered 9,263 men and 27 of- 
ficers, increasing to three regiments and almost 
23,000 by the end of Selim's reign (Shaw, 131-4). 
During the 1798-1801 confrontations with France in 
Egypt and against c Othman Paswan-oghlu [see 
paswan-oghlu] of Vidin [q.v.], the new troops acquit- 
ted themselves with some small successes, hampered 
always by the refusal of the Janissaries to serve with 
them. Reform in the navy was more successful, under 
the direction of Grand Admiral Kiicuk Huseyn 
Pasha, who saw to the construction of new warships 
and technical schools during the same period. 

To finance his reorganisation of the military and 


navy, SelTm established the Irad-i Diedid, a new at- 
tempt at a centralised budget, with revenues from 

vacant, absentee or poorly managed military fiefs of 
the long disfunctional limar system were seized and 
added to the new treasury. While significant revenue 
was generated, the provincial upheaval such measures 
induced exacerbated the growing disaffection of local 
notables, one of the significant causes of Selim's 

The War of the Triple Alliance (1799-1802) pitted 
the Ottomans against Napoleon, and forced a reluc- 
tant Sellm into agreements with Britain and Russia to 
counter the French invasion of Egypt. During these 
confrontations, the sultan was forced to rely on the 
private armies of local notables such as Ahmad Djaz- 
zar Pasha [q.v.\ in Acre [see c akka] and Sidon [see 
saydA], C A1T Pasha of Yanina and c Othman Paswan- 
oghlu, which both extended their power and increased 
the privations of the countryside, especially in the 
Balkans. The Treaty of Amiens in 1801, negotiated 
without an Ottoman presence, so angered the sultan 
that he signed a separate peace with France on 25 
June 1802, restoring that country to all its pre-war 
privileges and ignoring the question of war indem- 
nities which both Russia and Britain were de- 
Peace, however, meant the renewal of internal 
revolt, often encouraged by the empire's erstwhile 
allies, especially the Russians, who had made tremen- 
dous gains in the Treaty of Amiens. A serious revolt 
in Serbia against Janissary and the auxiliary Yamak 
abuses broke out in 1802, developing rapidly into a 
revolution under the leadership of Kara George after 
1804, and influencing much of the diplomatic man- 
oeuvring of the period. War between France and 
England broke out again in 1803, and intense 
diplomatic pressure by the resurrected Anglo-Russian 
alliance in Istanbul forced Sellm to sever relations 
with France in 1805. Further French victories over the 
European allies, however, persuaded Sellm to grant 
formal recognition to the emperor in 1806, and to 
declare war on Russia in December after the Tsar 
ordered the occupation of the Principalities and was 
continuing to support the Serbian rebellion. Britain 
sent warships through the Dardanelles [see canak- 
kal c e boghazi] to the capital in February, demand- 
ing the expulsion of Sebastiani, French ambassador to 
the Porte after 1805, and compliance with Russian 
demands vis-a-vis the Principalities. Selim's refusal to 
comply, and his orders to fortify the city and the 
Straits, let to the British withdrawal to Tenedos 
[see bozdja-ada], a last moment of victory and accord 
between the sultan and his people. The British fleet 
occupied Alexandria in 1807, but found that Muham- 
mad 'All Pasha [q.v.], governor of Egypt since 1805, 
had subdued the Mamluks [q.v.], forcing the British 
to withdraw. 

Selim's failure to create a broad coalition of sup- 
porters for his reform agenda, however, finally over 
whelmed him. A general call to arms for Nizam i 
Diedid troops in March 1805 had precipitated an ope